Bringing Project-Based Learning into Renewable and Sustainable Energy Education: A Case Study on the Development of the Electric Vehicle EOLO

Abstract

In recent years, there has been a growing interest in education for sustainable development (ESD). Although several national and international agencies, e.g., the UN or UNESCO, have promoted its deployment in higher education institutions, educators are still facing problems with how to articulate this type of education within the curriculum, allowing students to develop their technical and labor competencies, and soft skills as well. In this way, this study describes a methodology with Project-Based Learning in renewable and sustainable energies through the development of an electrical vehicle (EV) known as EOLO. This initiative arose from an industry-academia collaboration to develop the first Colombian EV with the support of solar and wind energy sources. Twelve engineering students participated in the development of the vehicle through a set of capstone projects over a year and a half with the support of two tutors (professors) and two engineers (technical staff) of the project. Additionally, two versions of EOLO with vertical and horizontal axis wind turbines were made with the cooperation of the students. The results evidence that the methodology helped to engage students, promoted meaningfully and situated learning through real-world problems in renewable energies, and fostered motivation and peer collaboration. Nonetheless, aspects such as the improvement of the communications channels, the revision of the complexity of the projects, the sense of community to achieve a common goal, or the tutoring and monitoring processes should be strengthened for further initiatives and/or active learning methodologies. In this sense, some challenges and recommendations that can help to develop methodologies that combine ESD and engineering are provided based on the experience in this study.

1. Introduction

During the last decade, the interest in Education for Sustainable Development (ESD) has increased considerably in the engineering curricula and it has inspired the reformulation of the role that engineers have in society and the impact that their products, designs, and services could have on the environment. Additionally, the Sustainable Development Goals (SDG) promoted by UNESCO [1,2,3,4], especially, SDG 4 (quality Education), SDG 7 (affordable and Clean Energy), SDG11 (sustainable cities and communities), and SDG13 (climate action) have accelerated this trend. Then, aspects such as the ability to determine the most appropriate renewable energy strategy for a community, the understanding of the importance of lifelong learning, the reduction of educational gaps and inequalities, or the capability to develop a reliable plan for sustainable energy production are announced as examples of key factors to achieving the SDGs [1,3].

Similarly, the need for responsible and eco-friendly engineering is continuously included and revisited in the assessment criteria of the accreditation agencies such as the Accreditation Board for Engineering and Technology (ABET) and the Engineering Council (UK-SPEC), which are the base to formulate engineering programs at the levels of undergraduate, master, and Ph.D. degrees in the US, the UK and in other higher education institutions (HEIs) around the world. For instance, the student outcomes in criterion 3 of the ABET (2020–2021) [5] indicate that graduate engineering students should have “an ability to recognize ethical and professional responsibilities in engineering situations and make informed judgments, which must consider the impact of engineering solutions in global, economic, environmental, and societal contexts”, as well as “an ability to apply engineering design to produce solutions that meet specified needs with consideration of public health, safety, and welfare, as well as global, cultural, social, environmental, and economic factors”. By the same token, the European Commission [6] states that the role of HEIs is to educate critical thinkers, problem solvers, and agents of change to be responsible citizens. For that, HEIs need to engage students in complex analysis of reality and create spaces to foster democratic participation and decision-making processes. Although around 1400 HEIs have endorsed, signed, and approved plenty of declarations about Sustainable Development (SD) from 1990 until now mainly in Europe and US [7,8], questions still remain about how engineering curricula can contribute to SD and how these curricula could keep students motivated and engaged. In particular, these issues have been reported in numerous studies in engineering education [9,10,11,12,13,14]. The authors claim in these studies for a need to understand engineering holistically and not only from the traditional technical point of view regarding SD. Also, as indicated in the study [9], in which SD was analyzed through the engineering curricula of 25 Spanish universities, the biggest identified challenge was the ensuring of SD competencies as a transversal initiative and not only from the isolated efforts of the academic staff to embed SD into the engineering curricula.

Nonetheless, as a counterpart, we also think that the contributions of Educators and Engineers from the classrooms are important to ensure effective inquiry and reflection about SD with the construction of methodologies, courses, and frameworks, among others, in this regard. With these contributions, the engineering curricula and policies in HEIs could be gradually changed and transformed to benefit the real integration and understanding of SD from the standpoint of students, teachers, and stakeholders in the education field. Then, the main purpose and motivation of this study were to create an educational methodology that responded to the learning needs of the students regarding sustainable energy education (solar and wind), and embedded and power systems in the design of the electric vehicle EOLO. This was a project in an industry–academia collaboration to design and deploy an electrical vehicle entirely produced in Colombia to contribute to the e-mobility in our country. In particular, this study was focused on the SDGs (4: Quality education, 7: Affordable and Clean Energy, and 9: Industry, Innovation, and Infrastructure). For the development of EOLO, (n = 12) students were engaged in the methodology of Project-Based Learning (PjBL or PBL) as part of a set of capstone projects. In addition, the 12 students received the accompaniment of two tutors and a continuous follow-up from the technical staff of the project EOLO for one year and a half. During this period, two versions of EOLO were constructed with the work, efforts, and cooperation of the students. The students were divided into small groups of two or three people with specific tasks and purposes to attain the proposed goals in EOLO and foster peer collaboration. The different problems addressed through the PjBL methodology allowed the students to inquire, design, develop, and deploy solutions, employing the skills and competencies acquired along their learning process. Although it seems that the number of students is small, this was because the amount of projects in EOLO was limited and one feature of the PjBL methodology is the work usually in small groups to achieve the proposed goals (educational and technical).

To evaluate the methodology and the experience, a survey with 19 questions was provided to the students. Additionally, a second survey was produced with 13 questions for the two technical leaders of EOLO in order to complement the standpoint of the students and identify recommendations and aspects to improve in the methodology. With the previous approach, we search to address the lack of methodologies that help to embed the SDGs in the curricula of engineering, especially those mentioned above. Also, we want to describe in-depth the details and results of the methodology, which can benefit other educators in the construction and/or incorporation of active learning methodologies in the areas of renewable energy, sustainability, embedded systems, and electrical vehicles, among others. It is worth mentioning that the results presented here expand on those presented in the conference paper [15].

This paper is divided into the following sections. Section 2 shows the related works. Section 3 describes the details of the EOLO project, participants, instruments, and the methodology including some technical factors. Section 4 outlines the results and discussion of the methodology from the standpoint of the students, tutors, and main staff of the EOLO project, including some representative technical results. Section 5 exposes several educational challenges and recommendations that were identified in the process. Finally, Section 6 depicts the limitations, further work, and conclusions of this study.

2. Related Works

2.1. Educational Works

Several educational methodologies have been described to foster ESD, mainly in HEIs. For instance, Cebrián et al. [12] expose four student-centered methodologies, namely, project or problem based-learning, case studies, simulations, and cooperative inquiries. The authors employed both a literature review and an expert analysis to determine the importance of each one of the previous methodologies in the process of ESD. Authors conclude that the smart classrooms, that is, the learning settings that use the methodologies indicated above can foster collaboration, knowledge sharing, and successful interaction between students and teachers. Similarly, the authors in [16] proposed a project-based learning methodology for non-technical students with different laboratory experiments in the fields of control and actuator technologies and e-mobility. The authors suggested that the methodology can help to bring to life the discourse about ESD and it allowed its incorporation inside interdisciplinary projects in electrical engineering. Also, a methodology for the development of eco-design and eco-efficiency competencies in engineering is presented in [13]. In the methodology participated 49 students during 2018–2019 who performed three assignment stages: Identifying the product and fabrication processes, eco-design analysis of the product and eco-efficiency evaluation, and rethinking the product to decrease its environmental impact and promote circularity. In the conclusion is highlighted the importance of the interaction between students and industrial agents to create new ways for collaborative projects. In the same way, Curiel-Ramirez et al. [17] propose a new curriculum for automotive engineering education. Through the analysis of the topics regarding advanced driver-assistance systems (ADAS), product design, and chassis, among others, in several universities around the world, the researchers identified the key factors for the new curricula.

Additionally, an approach focused on inquiry-based learning for renewable energy sources (RES) is described in [18]. The approach helped students with low academic performance and it engages them in the topics of RES through problem-solving activities that address some issues in the current renewable energy panorama. In addition to this approach, the authors in the work [14] discuss an eight-year experience teaching aerodynamics, mechanics, and hydraulics, which were aligned with the SDGs, critical thinking, and problem-based learning. The authors depict the different phases such as definition and planning, monitoring and execution, assessment, and group and learning activities. The results of the methodology evidenced that the activities generated reflective thinking and a cooperative learning environment about four SDGs (SDG4: Quality education, SDG13: Climate change, SDG7: Affordable and clean energy, SDG 6: Clean water and sanitation). Alternatively, an educational strategy for ESD through a kit for energy, environment, and sustainability in Guinea-Bissau is proposed in [19]. The authors indicate the experience to use the kit called KEAS with several teacher trainers in the Guinea-Bissau School of Education. The authors mention several principles for the design of the kit such as being culturally appropriate, user-friendly for students and teachers, low-cost, or sensitive to gender issues. In addition, with the kit, teachers can explore practical work in science classes, helping to enhance the learning process of the students.

Although the previous methodologies are successful cases of the incorporation of ESD principles in an international perspective, in the Colombian context, however, there are few examples of methodologies in engineering education that help to understand the implications of the ESD both into the curricula and the educational pathway of the students. Thus, some examples of national policies and related works have been exposed in the studies [20,21,22,23], but still in some of them it is not clear how the SDGs can be integrated properly into the technical competences or abilities that engineering students should develop. This may be due in part to the fact that the SDGs are viewed in a general way in courses that are not an active part of the engineering curricula. For example, in these courses, students can address environmental care through recycling activities, water care, sanitation, fundamentals of SGDs, etc., but one issue here is how to properly integrate these concepts into the main corpus of technical abilities or competencies that one student must have in engineering and more specifically in this work, in electronic or electrical engineering or technology. Some strategies have been elicited to overcome this issue in higher education (HE), which include training through lectures, involving students, individual interaction, project-based and inquiry-based learning, and promoting personal and collaborative learning along the sustainability context [12,24,25]. In addition, HEIs should be leaders in academic changes that include the ability to reformulate curricula that respond to the SDGs through initiatives such as linking ESD pedagogies to special literacies, e.g., in science, reading, or mathematics, digesting new sustainable thinking and practice in the industry/profession, and rethinking quality learning outcomes under the lens of ESD [26].

Complementarily, this study was shaped by a work in progress (WIP) presented to the IEEE Frontiers in Education Conference (FIE) [15]. In this preliminary study, we exposed the results of the students’ perceptions regarding the methodology forwarded in the first version of EOLO. At this point, it is important to mention that two versions of the EV—EOLO V${}_1$ and EOLO Pro—were built with the support of the students during the years 2017–2020. The conclusions of this first work show that students were engaged in the methodology and learned concepts about renewable energy sources, CAN-Bus communications, electrical wiring for EVs, and embedded programming. For this reason, e.g., the 75% of the students stated that the projects increased their motivation for learning concepts and applying them in the deployment of their solutions in the EV. Similarly, at this point, the students manifested the need for reinforcing the tutoring process with the teachers and technical staff of EOLO. These suggestions were taken into account in the construction of the second version of EOLO called EOLO Pro.

2.2. Technical Works

At the technical level, some related works were deemed, even though the interest in the current study is not technical but instead educational. For instance, in the study [27], the author describes a hybrid vehicle that combines a photovoltaic (PV) module with a micro wind turbine. During sunny days, the PV module and the turbine can add 19.2 km of extra autonomy to the vehicle. Nonetheless, this autonomy is achieved at the cruise speed of 121 km/h. A second experimental work is presented in [28]. In this work, the author simulates the behavior of a car with a vertical axis turbine, employing computational fluid dynamics (Autodesk CFD). The author emphasizes the shape of the turbine blades as a factor that influences the pressure distribution on the car and can cause fatigue in the vehicle’s parts in the front and rear. Another experimental study about the possibility to install a wind turbine in an EV is depicted in [29]. In the simulations, the wind turbine with a voltage generator can provide an electrical power of 3.26 kW when the car is moving at a speed of 120 km/h. Finally, in this work, the authors recommend further work to identify how the power generated could be increased. In reference [30] is shown a rapid prototyping technique with 3D printing to create a wind turbine focused on vehicles and domestic power generation. The study exposes the feasibility to implement and deploy a vertical wind turbine with PLA with biodegradable properties and a duration of 20 years. Moreover, the simulations show that the turbine can be used in vehicles up to the speed of 80 km/h. Previous works are meaningful examples of the feasibility to create and install a wind turbine in a vehicle for electrical power generation. Nevertheless, there are some points to improve in these works. The first one is the need for the vehicle to reach the cruise speed of 120 km/h as claimed in the studies [27,29], which could not be optimal for urban traffic. The second consideration is that many of these works employ simulations as proof-of-concept and do not show real implementations in EVs. Searching to address this technical gap is just how the project EOLO arose. The vehicle’s concept was focused on the construction of a wind turbine that was feasible for EVs, and on the other hand, on features of the EV to gain autonomy with the drop-off of the wind resistance thereof, e.g., creating different wind ducts in the front part of the EV. The tests of the vertical wind turbine installed in the vehicle, which were performed directly in a racetrack of 667 m long, completing 110 laps, can be consulted in the study [31]. The main results in the tests indicate that the wind turbine can provide electrical power over the vehicle’s speed of 50 km/h, and allows increasing the vehicle’s autonomy between 4% to 8% considering that the maximum boundary of power generated was 1289.59 W and the vehicle consumption was 14,529.9 W. For researchers interested in the performed technical simulations and tests (see work [31]).

3. Context and Method

3.1. EOLO Project

The EOLO project arose in 2015 as an initiative to create an electrical vehicle 100% designed and produced in Colombia. The project was released by the Corporación Industrial Minuto de Dios (CIMD), a Colombian entity that aims to serve the entrepreneurial and industrial development and income generation, through the protection and generation of employment, the dynamization of entrepreneurship and training for productive qualification [32]. EOLO was the first electric vehicle (EV) developed in Colombia with a proprietary wind charging mechanism through Savonious turbines [33]. These devices can be installed in small places, e.g., in the front part of the vehicle, and allow us to propel voltage generators to charge a set of installed lithium-ion batteries. The outcome of the incorporation of these devices is an increase in the vehicle’s autonomy between 4–8% in the EOLOs case [31]. Two EOLO versions known as EOLO $V_1$ and EOLO Pro were designed and constructed between 2017–2020, and their appearances are depicted in Figure 1. The first version of EOLO has a vertical axis wind turbine (VAWT), meanwhile the second one has a horizontal axis wind turbine (HAWT) to save space and weight in the EV. Additional information about the vehicle can be found in the references [31,34,35,36,37]. The main technical details of both versions can be seen in

Table 1. Technical parameters of both EOLO’s versions. Notice that in both versions of EOLO the same electric motor was employed.

Parameter	EOLO${\mathit{V}}_1$	EOLO Pro
1. Motor reference and power (HP)	HPEVS AC51-88 HP	HPEVS AC51-88 HP
2. Torque (ft-lbs)	108 ft-lbs	108 ft-lbs
3. Power system voltage (V)	144 V	144 V
4. Battery Type	12 V-LiFePO${}_4$	12 V-LiFePO${}_4$
5. Number of batteries	48	48
6. Max. speed ($Km/h$)	120 km/h	120 km/h
7. Charging time (hrs.)	8 h (120 V)–5 h (220 V)	8 h (120 V)–5 h (220 V)
8. Weight (Kg)	1200 Kg	800 Kg

.

As described, the Savonious turbines were installed vertically as Figure 1a,b depict for EOLO $V_1$. Concerning EOLO Pro, the turbines were redesigned for searching less weight and better improvement of the autonomy of the vehicle, which is still under development. In this way, turbines were installed horizontally. In addition, for this last version, a second photovoltaic system with a nominal capacity of 118 W was installed on the vehicle’s roof and in its rear part. Six panels were installed with a total capacity of 100 W (12 V–7.5 A) for three of these in the roof, and 18 W (12 V–1.5 A) for three of these in the rear part of the EV as Figure 1d depicts. The goal of both systems (solar and wind) was to improve the vehicle’s autonomy by a maximum of approximately 11% (8–10% wind + 0.6–0.81% solar) considering the power vehicle’s consumption mentioned above. Nonetheless, this improvement in autonomy is currently under revision and development. In this way, for the design, monitoring, and deployment of the energy and the electric systems for both versions of EOLO that included, e.g., lighting, communication, and control systems, the CIMD made an industry-academia collaboration with the university Corporación Universitaria Minuto de Dios-UNIMINUTO in 2018. Following this alliance, two professors and twelve students from the engineering faculty were engaged in the project. Thus, according to several meetings held between the group of professors and the technical staff of the EOLO project to gather technical requirements, six projects were generated, which will be discussed in the next sections.

3.2. Context and Participants

The students and professors who participated in the project belong to the program of Technology in Electronics at the Colombian university Corporación Universitaria Minuto de Dios-UNIMINUTO. In Colombia, technological careers are understood as types of programs focused on labor fields, with a duration between two and three years. Most of the subjects established in the curricula of those programs are focused on the necessary abilities that a determined industrial or productive sector requires, e.g., in some programs of technology in electronics, their emphases include telecommunications, industrial communications, automation, or bioengineering [38]. The students had a range of ages between 17–35 years old and they were between the fourth and sixth semesters of their careers in Technology in Electronics. Hence, the students already possessed knowledge of embedded systems, circuits, or the fundamentals of power electronics. Nonetheless, as we shall see, several additional technical concepts were developed or reinforced with the group of students in these areas. Furthermore, eight students worked during the day and took their classes in a night schedule (6:00 p.m.–10:00 p.m.) and typically performed the different tasks in the project on Saturdays. Besides, two engineers responsible for the EOLO project monitored the students throughout the process of the accomplishment of the proposed technical tasks. The students invested approximately between 8–12 h per week on the tasks assigned to each project.

3.3. Educational Objectives

Three educational objectives (EOs) were constructed for the different activities performed by the students. These are listed as follows:

• EO1: Reinforce and strengthen the learning about sustainable energy sources (wind and solar), embedded, and power systems with the design and deployment of technical solutions for the projects proposed in EOLO.
• EO2: Motivate and engage the students to create solutions eco-friendly and eco-responsible from the technical standpoint regarding e-mobility and sustainable energy sources.
• EO3: Promote teamwork and the development of soft skills in the students that help to fulfill the goals and aims of EOLO, which includes enhancing the educational process of the students.

Our primary objective was to provide high-quality education to our students with a representative project in e-mobility that allows them to put into practice the knowledge and learning developed in their educational pathway in the areas of sustainable energy sources, embedded, and power systems. In this sense, this aim is aligned directly with the (SDG4: Quality education), but the different actions of the students in EOLO also are aligned with the (SDG7: Affordable and clean energy), and (SDG13: Climate change). In addition, in the normal curriculum of the career of Technology in Electronics, the topics related to sustainable energy systems are not encompassed. Therefore, throughout the project, the students reviewed these topics and applied them with hands-on activities, which is a key point in engineering and technology education.

3.4. Description of The Capstone Projects

As described, six projects were outlined according to the technical requirements in EOLO. The projects combine embedded systems, handling of power electronics, CAN-Bus communications, Human Machine Interface (HMI), and battery balancing. The goal of the projects in EOLO was to employ the power generated by the wind turbine and the solar panels in several secondary systems like the previous ones without consuming the power of the primary system that provides power to the electric motor, whose nominal capability is 15 KW. Moreover, the remaining generated electric power can be used to charge the main battery system of the EV.

The projects were continuously revisited in complexity and relevance by the professors and main technical staff involved in the project before assigning them to the students.

Table 2. List of formulated projects for EOLO.

Project	Description
1. Balancing batteries	Balancing system for at least 4 Lithium Iron Phosphate batteries (LPF) ($LiFeP{O}_4$). These batteries serve as the power supply for the electric engine of the EV.
2. Electric windows and mirrors	Current sensing and control system for the mirrors and the electric windows.
3. Redesign of electric wiring	Redesign of the electric wiring according to the technical requirements.
4. Lighting system front and rear	Lighting system with LED lamps controlled by CAN-Bus protocol.
5. Electric mechanism for vehicle’s doors	Electric mechanism to open and close the vehicle’s doors.
6. Driver-focused Human Machine Interface (HMI)	HMI with a touch screen that supports functions such as tachometer, battery status, command buttons for light intensity (high, medium, low, strobe) and stationary lights.

shows the list of these projects with their technical description. The formulated projects responded to the following requirements:

• Reduction of the power consumption of the lighting system. The first version of this system had a consumption of 300 W with halogen lamps, which made it unfeasible.
• A balancing system for the batteries. Without this system, the batteries could be damaged. The cost of each battery is approximately USD 250 and the vehicle needs 48 of these.
• A HMI system since the first version of EOLO did not have it. The system provides different functions to the driver in order to control some features of the vehicle. If it is possible, the system should have a touch screen for some of these functions.
• A system with CAN-Bus protocol (typically used in automotive engineering) to communicate the different units or nodes in the vehicle and reduce electric wiring.
• Redesign the electric wiring according to the normative in vehicles [39] to optimize the number of conductors needed in the vehicle.
• Introduction of the Savonious turbine and the solar PV panels in the recharging system of EOLO pro.

With the previous requirements, at the technical level, the projects were oriented towards the architecture presented in Figure 2 for the electric system of the vehicle. Here, by now, the function of turning on the reverse lights is not available. The systems presented in this figure are interfaced to a second battery independent of the primary set of batteries of the EV, which is charged with the wind turbine and the solar panels. In the schematic of Figure 2, the devices and components were selected initially by the professors and engineers of the EOLO project. The hardware devices such as dsPICs, which are Digital Signal Controllers (DSCs) [40] for the design of the different nodes in the vehicle, including the Electronic Control Unit (ECU), the touch screen with the FT900Q microcontroller for the HMI [41,42], or the low-cost pulse width modulation (PWM) controller FL7760 [43] that controls the intensity of the front LED lamps, were selected according to parameters such as energy consumption, price, robustness, scalability, easiness to program, and availability of educational and technical resources to learn and understand their functioning. Notice that the different nodes, e.g., the ECU and the nodes for the front and rear lights, are communicated through a CAN-Bus protocol. This type of communication is not normally viewed in the courses on embedded systems in our curricula, for this reason, the fundamentals of its functioning and its importance in automotive systems were taught to all students who participated in the methodology. In parallel, other additional concepts in renewable energy, photovoltaic systems, embedded programming of DSCs, and functioning of electrical vehicles were introduced to the students since these were employed transversely in all projects in Table 2. Additionally, these concepts were reinforced by the tutors in the project, and the students had the necessary educational materials.

3.5. Educational Materials

As a technical requirement, the students must incorporate dsPIC devices in the creation of each node. We selected these devices due to some features such as processor speed, power consumption, and the number of protocols including CAN-Bus, or remapping pins. The programming of the dsPIC devices is not a current topic in our curricula. Then, the professors decided to create some utilities and educational materials based on prior works. For instance, for the programming of the dsPICs, the tutors created a utility through block-based programming called DSCBlocks. Through this application, the students can create codes using graphical blocks and can directly program the dsPIC with them to handle different peripherals such as (CAN-Bus, UART, PWM) or ports thereof. The students at any moment can see the respective code for the blocks in C language for the architecture of the dsPIC. Additionally, the supporting educational materials to understand the 16-bit dsPIC architecture were provided to the students before the stage of design. The overall appearances of the application and the educational materials are depicted in Figure 3 and Figure 4, respectively. Detailed information about the application DSCBlocks can be found in the reference [44], and in the related study [45]. The e-learning tool Xerte [46] was employed as a tool for the construction of the educational materials that started with the concepts of the dsPIC architecture and finished with the features of the application DSCBlocks. Xerte provides a complete environment to develop materials with interactive resources such as videos, quizzes, games, or web pages that can accompany the learning process of the students. Other educational materials in solar photovoltaic panels, battery management, and electric vehicles were provided to the students timely with Xerte.

3.6. Methodology

The educational methodology was structured according to the guidelines of the active learning methodology Project-Based Learning (PBL or PjBL). PjBL is defined in terms of an interdisciplinary, student-centered, and constructivist approach that enables collaboration and reflection within real-world practices [47,48,49]. Among the advantages of PjBL are highlighted learning of students or teachers, collaboration and sense of community, motivation, student-centered learning, and versatility for education [47]. According to Dumitrache et al. [50], the projects should be formulated considering aspects such as relevance for the discipline, complexity, attractiveness, and simplicity. Also, the students must know how they will be evaluated and the evaluation criteria. Then, several capstone projects were generated between the technical staff of EOLO and professors since their inclusion has been demonstrated to be a catalyst of meaningful learning, peer collaboration, and motivation in engineering because students should use the learning and knowledge developed along their education pathway [51]. Capstone projects and courses constitute the culminating experience in engineering curricula [52]. Thus, in this case, the methodology in these projects had the following overall stages:

• Project formulation.
• Project selection.
• Project design, development and deployment.
• Project debugging.
• Presentation of final results.

Each one of these stages will be explained in the following subsections. It is important to indicate that the stages of project design, development, deployment, and debugging had a continuous follow-up and feedback by the professors and technical staff of EOLO. In this way, any design created by the students was implemented and deployed in both versions of EOLO. If an error or malfunction was detected, the feedback of the staff mentioned, allowed the students to return to the stage of design to check the problem and found the root of this to fix it, and again deploy the solution.

3.6.1. Stage 1: Project Formulation

The projects were formulated according to the technical requirements of EOLO. As described, multiple meetings were held to formulate and coordinate the projects between the professors and the engineers in EOLO. Since the requirements were defined previously by the technical leaders of EOLO, firstly, professors evaluated the pertinence, relevance, and complexity of these requirements for the students according to an examination of the required competencies in the function of the curricula. Secondly, with this examination, the professors determined other additional concepts that the students should address in the development of the technical tasks. Then, each project of Table 2 was formulated and conditioned according to the abilities that the students must display under the lens of the curricula. In parallel, as shown, 12 students that manifested their intention to contribute to the EOLO project were selected to participate in it.

3.6.2. Stage 2: Project Selection

The students were asked to choose the project more convenient under their abilities. Moreover, the students selected their peers in the project and worked in groups of two or three people. Only a group of one person was allowed since the tasks and complexity of this project were enough for a student. Also, the students had several meetings with the EOLO staff to know the technical requirements and expand the perspective of the projects.

3.6.3. Stage 3: Project Design, Development and Deployment

In this stage, the students started with the design of their solutions for each project. In time slots of 20 days, the students exposed their advancements to the professors and main engineers of EOLO in a short document and presentation. Overall, this stage had the steps described in Figure 5, which started with the activity of brainstorming and finished with the implementation and testing of the solutions proposed by the students.

In the brainstorming part, the students by group proposed several solutions for their projects. Afterwards, these solutions were reviewed and evaluated initially by the group of professors/tutors that accompanied the process. Then, based on the feedback received, the students chose the most suitable solution considering aspects such as cost, the feasibility of implementation, energy consumption, and the availability of hardware devices in local stores.

With the solution proposed and the evaluation of its feasibility, the students started with the process of concept comprehension and design. Regarding concept comprehension, the students understood the additional concepts needed to make the designs of their solutions. In the design were deemed technical aspects such as power consumption, power saving modes for the DSCs, processing speed, data analysis of current and voltage for the solar PV panels and the wind generator, distance between the nodes to perform the CAN-Bus communication, the electric noise, the distribution of the HMI, among others.

Once the designs were concluded, the students started with the step of prototyping. In this, devices such as a 3D printer, Arduino, some development boards for DSCs, and the touch screens for the HMI were used to speed up this step and to test the prototypes quickly in the versions of EOLO. Some examples of tasks made by the students were: Programming basic commands (On-Off) in dsPIC, reading and writing data in the Universal Asynchronous Receiver-Transmitter (UART) peripheral, electric modeling of mirrors and windows, temperature sensing in Arduino, modeling consumption of led lamps, configuration of basic buttons in screens, PWM control, etc. In the second part of the prototyping stage, the students improved their designs and they built the respective printed circuit boards (PCBs) and their housings in a 3D printer in the function of the technical requirements.

The students tested their prototypes directly in the versions of EOLO. Nonetheless, the EOLO $V_1$ was transformed in a laboratory as the students performed the different tests, and these prototypes were quickly adapted to the EOLO Pro version. The advancement in the prototyping and its implementation was continuously monitored by the tutors and engineers of EOLO. Over 1250 h of tests were employed in the different nodes that were developed in this stage in continuous work with the students.

3.6.4. Stage 4: Project Debugging

Any error presented in the design and implementation was analyzed jointly by students, tutors, and engineers to find the source of the failure and provide a timely solution to the problem. Then, students made the corrections to their designs, executing the steps in Figure 5 again.

3.6.5. Stage 5: Presentation of Final Results

When the student finished their designs and their implementation in both versions of EOLO, they exposed their results in a report and presentation before the community and some academic peers. With this task, the students completed their training within the EOLO project. Nonetheless, as indicated, students submitted every 20 days a brief report with their activities and advances. Some figures that illustrate the previous stages are depicted in Figure 6.

3.7. Instruments

In order to assess the methodology, a survey with 19 questions (13 closed-ended, 6 open-ended) was applied to the students. The survey sought to know the perception of the students concerning learning, motivation, problems, and recommendations with the methodology. The students’ survey was expanded in the present study regarding its first version whose results are consigned in the prior study [15]. Also, for this new instrument, the closed-ended questions were complemented and divided into three sections, the first one with 8 questions (Q1–Q8) and Cronbach’s alpha ($\alpha$) = 0.88, ranged from (1) Strongly disagree to (5) Strongly agree. The second one with 3 questions (Q9–Q11), ranged from (1) Very poor to (5) Excellent. One question (Q12) assessed the difficulty of each project in the range of (1) Very easy to (5) Very difficult. In the last one (Q13), the students provided a global evaluation of the methodology in the range of 0–5. The value of ($\alpha$) was increased for the first version of the survey consigned in [15] from ($\alpha$) = 0.71 to ($\alpha$) = 0.88 for the current survey, which demonstrates the reliability of the instrument [53].

Also, as described, a second survey was performed to know the perceptions of the main technical leaders of EOLO regarding the process followed by the students and its impact on the project, which complements the results posed in the prior study [15]. The survey had 13 questions (8 closed-ended and 5 open-ended), 5 closed-ended questions on a 5 points Likert scale (Q1–Q5), 2 questions in the range (1) Very poor to (5) Excellent (Q6–Q7), one in the range 0–5 (Q8), which evaluated the overall performance of the students in EOLO, and the other five of the open-ended type. In some cases, given the number of questions or respondents, it was not possible to calculate the value of ($\alpha$). The questions in the surveys respond to the EOs indicated in Section 3.3 and the need to know the perceptions of both the students and the technical staff of EOLO regarding the process and the methodology addressed. Although the instruments are self-reported, these provide us with important feedback about the methodology, the positive points, and aspects to improve for further methodologies and/or educational projects.

3.8. Analysis

To analyze the educational data, a mixed approach was adopted through embedded design, which is employed in engineering education studies [54]. In this, the quantitative and qualitative data collected with the mentioned instruments are analyzed in parallel to find the outcomes of the study and contrast them with the theoretical concepts or constructs. For the analysis of quantitative data of the surveys, e.g., mean (M) and standard deviation (SD), the software IBM SPSS v.27 was used, while the Computer Assisted/Aided Qualitative Data Analysis Software (CAQDAS) NVIVO v.12 was employed for the collecting and analysis of qualitative data thereof. Complementary, the main technical results of some projects are exposed, which include the lighting system, the CAN-Bus communication, the HMI, and the PV solar system.

4. Results and Discussion

4.1. Educational Results

Table 3. Descriptive statistics for the closed-ended questions in the student survey, $(n=12)$.

Question (Q)	M	SD
Q1. With the methodology used, Did you use in the design and implementation of your project, the knowledge and/or technical skills developed during your learning process?	4.17	1.19
Q2. Do you consider that the project developed was clear in its formulation and objectives?	3.92	1.16
Q3. Do you think that the methodology contributed to the development of your learning?	3.58	1.44
Q4. During the project, was your motivation increased to create the solution to the given problem?	3.83	1.4
Q5. Do you think that the learning in the development of your project will be used in your professional practice?	3.58	1.16
Q6. Do you think that the methodology allowed teamwork and collaboration, which helped to achieve the project’s goals?	4.33	0.65
Q7. Were the tutors available to solve the different doubts and issues in your project?	4.5	0.67
Q8. Was it difficult to obtain the materials and equipment required for the design, development, and implementation of the solution to the problem posed in the EOLO project?	3.58	1.44
Q9. Please, assess your motivation and willingness in the developed activities and tasks in EOLO	4.42	0.51
Q10. Please, assess your performance working in team in EOLO	4.42	0.51
Q11. Please, evaluate the support of the tutors in relation to the EOLO project	4.75	0.45
Q12. What level of complexity do you consider the project developed at EOLO to have had?	3.5	0.52
Q13. Finally, taking into account the previous elements, please, assess the methodology in a range of 1–5.	4.33	0.78

describes the descriptive statistics for the closed-ended questions in the student survey. Similarly, a chart with the distribution of the questions (Q1–Q8) is depicted in Figure 7.

Here, for example, 75% of the students considered that the different tasks performed in EOLO allowed the development of knowledge and learning (see Q3). This learning will be used by 58.3% of the students in their jobs (see Q5). Some of our students currently work in other fields of electronics different from the EVs and this could be a reason for the result. Despite the results in Q4 (M = 3.83, SD = 1.4), 75% of the students indicated that their motivation in the projects increased through the methodology, while 83.33% of the students in (Q1) employed the technical skills that were developed throughout their educational process in the different tasks and technical requirements in EOLO. In the same line, the students established that the teamwork helped the fulfillment of the different purposes and goals in each proposal, (see Q6: M = 4.33, SD = 0.65), which corroborates the coherence between the methodology and the collaborative learning. Even, the students manifested that the methodology and projects foster entrepreneurship, motivation, and teamwork as well as their problem-solving skills.

The students mentioned that they learned and applied concepts in the areas of programming, embedded systems, PCB design, battery measurement and control, design of 3D models in AutoCAD and CorelDraw, and photovoltaic systems, among others. These results are aligned with prior studies related to ESD and PBL. For instance, in the courses that employed this combination, students manifested an enhancement of their learning process and the possibility to apply their knowledge in relevant real-world problems [14,18,55]. This is also aligned with the results obtained in the prior work with EOLO $V_1$ [15]. For instance, in the survey followed in this prior study, the students classified the level of difficulty of each project between difficult (50%) and moderate (50%), and 58.3% of the students established that the generated apprenticeship in the project will serve in their future labor life. Similarly, 83.33% of the students employed the technical skills developed throughout their educational process.

In Q8, the students indicated the difficulty to obtain materials and equipment. The result reveals a problem because some hardware devices were imported, increasing the time of development in each project. Thus, we tried to use hardware devices that were found in the local stores while the selected components arrived at each project. However, several students stated that it would have been a good idea to provide them with all the necessary materials for their projects. Under this issue, the original purpose in some projects was changed to meet the time requirements of the project, and the students evaluated this issue with the result of Q2 (M = 3.92, SD = 1.16). In this respect, some students (S3, S4) indicated some aspects to improve:

S3. Establish in advance the materials and products required since it takes a long time for these materials to be delivered and used for the development of the project.

S4. Initially give a concrete and specific formulation of the project, making clear the work to be developed and its scope.

Therefore, as an improvement action, it is necessary to reduce the time required for the purchase of materials and the administrative procedures at the university.

The students classified the level of difficulty of each project between difficult (50%) and moderate (50%), which means that the problems had a correct level of complexity and relevance in the engineering field (see Q12). This result is expected in methodologies that employ PBL since a requirement is the level of complexity and relevance of the projects created by the students [47,48].

Unfortunately, not all the tutors were available to the students in the project for time reasons. This fact could have affected the perception of other questions in the survey. Hence, this was a difficulty in the methodology and the students mentioned and evaluated this trouble in the survey. In this way, the tutor’s accompaniment represented a critical point in the development of EOLO. From our perception, we believe it is important to improve the communication channels with the students. In this respect, two students (S1, S2) commented:

S1. I think that is a project with a vision and it has great potential, however, there is a lack of support from one tutor in the project. But also, I consider that the methodology functions well and it provides the opportunity for self-learning.

S2. It is needed to assign tutors who are engaged with the project, otherwise, the execution time for the project can be affected.

Nonetheless, to solve this issue, some tutors helped with the different doubts and inquiries of the students regardless of the time assigned for the project. For this reason, about 91.6% of the students in Q11 (M = 4.75, SD = 0.45) evaluated the support of the tutors between good and excellent. Concerning this, two students (S3, S4) mentioned:

S3. There was great support from the tutors and engineers of EOLO.

S4. EOLO is an excellent project to learn and it has a broad field of knowledge application to explore and research.

Despite the mentioned problems, the methodology has a good acceptance among the students (M = 4.33, SD= 0.78). Most of the students indicated that the methodology gave them the possibility to exchange knowledge and experiences with engineers and teammates, allowing to enhance their learning process.

As for the EOLO Engineers,

Table 4. Descriptive statistics for the EOLO engineer’s survey, $(n=2)$.

Question (Q)	M
Q1. Do you think that the problems and solutions proposed in EOLO contributed to student learning?	5.0
Q2. Did the students accept the advice and feedback suggested to obtain the solution to the problems posed?	5.0
Q3. Do you think the students performed a correct implementation of the problem solution?	4.0
Q4. Do you think that the different learning obtained in the development of the projects by the students in EOLO will be used in their professional practice?	5.0
Q5. Do you believe that the methodology used in EOLO allowed teamwork that helped to achieve the project’s goals?	5
Q6. Please evaluate the performance of students working as a team on the EOLO project.	4.5
Q7. Please evaluate the motivation of the students in the development of the project in EOLO.	4.5
Q8. Taking into account the above elements, on a scale of 1 to 5, evaluate the methodology followed by the students in the EOLO project.	4.5

describes the results for the closed-ended questions in the survey applied to them. In this case, the SD was not calculated for each question due to the number of respondents. So, the engineers mentioned the following outstanding aspects of the students in the methodology:

• Interest to be participants of the EOLO Project.
• Compromise with their responsibilities.
• Self-demanding with their activities.
• Curiosity in all projects at the mechanical and electrical levels.
• Cooperation and disposition towards the activities outside their projects.

The same engineers (E1, E2) evaluated the performance of the students in a range of (0) to (5) with an overall value of 4.5 (see Q8). As difficulties, recommendations, and comments, they indicated:

Difficulties:

E1. The students’ time availability for the development of their activities and the socialization meetings.

E2. One difficulty may be to adjust the students’ time with their advising teachers and the EOLO team. Also, due to the time constraints of the students, it is sometimes difficult for them to attend all the meetings proposed to socialize the planning of the projects as well as the progress, which directly affects the schedule of activities.

Recommendations:

E1. (1). To be able to generate more and better communication channels between students, tutors, and engineers of the EOLO project to achieve better control of developed activities and to coordinate more effectively the fulfillment of the activities. (2). More time available from the university so that tutors can give the necessary attention to their students. (3). More Documentation of progress made by the students in their respective projects.

E2. One recommendation may be to adjust the purchasing processes for materials, supplies, and equipment required for the project since the deadlines are very short and a delay in the purchasing process delays the projects.

Comments:

E1. At a general level, it has always been proposed that the university should bring its students closer to the industry and the EOLO project has been a research niche to involve students in industrial issues and more specifically in electric mobility, which is one of the new technologies that are beginning to position themselves worldwide.

E2. The problems or challenges faced by the students are based on real requirements, where they were able to demonstrate their capabilities, providing solutions tailored to the technical needs.

Previous comments and outcomes illustrate that is possible to create meaningful projects that enhance the learning process of the students through industry-academia collaboration. Nevertheless, it is necessary to improve the communication channels and reduce the administrative and purchasing times that affect the execution of the projects. As described, the students will employ the knowledge and learning created in the project for their labor life, which is an important consideration in the methodology with PjBL. With the industry–academia collaboration, the learning can be more situated to the real-world problems, in this case, in the area of renewable and sustainable energy, and EVs, contributing from the technical and educational perspectives to the achievement of the SDGs. The students directly put their knowledge to the advance of their projects with different hands-on activities, which is a clear differentiating factor in active learning methodologies. There are numerous success studies in the engineering context, mainly in software engineering, that illustrate the advantages that this type of collaboration (industry-academia) has on students [56,57,58], but also the challenges, in special with the relevance of the projects, the lack or drop of interest, the communication issues, and the monitoring process [59,60] that we have also experienced. Moreover, the industry–academia alliance can enable the development of high-quality personnel and motivate researchers to grow professionally within organizations [61].

Under these elements, our methodology integrated the components depicted in Figure 8. These elements can help to create meaningful learning with concrete and situated experiences that are included in the engineering curriculum. Finally, the previous results evidence that the OEs posed were accomplished and benefited the educational process of the students.

4.2. Technical Results

4.2.1. Overall Results

At the technical level, the most representative results for the systems developed by the students were:

• Reduction of wire gauge from AWG (14) to AWG (18), which decreases the costs relating to the electrical wiring.
• Robustness in the vehicle communication through CAN-Bus protocol.
• Incorporation of a Real-Time Operating System (RTOS) to prevent undesired delays in the functions of the hand-held control and in the screens commands.
• Reduction of 72.3% in the power consumption in contrast with the first electric system in EOLO $V_1$. The first system had a power consumption of 318.5 W and the new system 88.2 W.
• Low-cost user interface that gathers both a tachometer screen with vehicle variables such as speed, voltage, and current, and a second touch screen that provides the control of functions, improving the visual experience of the user.

In particular, the electric power reduction was analyzed with its measurement for each one of the subsystems made by the students vs. the previous one before their intervention as

Table 5. Comparison of power consumption of the new electric system developed by the students vs. the previous one ($V_{battery}$ = 12.7 V). The nodes are those described in Figure 2.

Component/Subsystem	Power Consumption (First System)	Power Consumption (New System)
High Lights	129.1 W	30.6 W
Medium Lights	-	18.9 W
Low Lights	88 W	12.4 W
Turn Signals	18.9 W	5.5 W
Stationary Lights	32.7 W	8.7 W
Brake Light	49.9 W	2.4 W
Driver Interface	-	3.43 W
Nodes	-	6.4 W
Total	318.5 W	88.2 W

shows. In this case, the user interface is composed of the touch and tachometer screens for the driver as Figure 6g depicts.

In EOLO $V_1$ shown in Figure 1a,b, the lighting system was composed of halogen lamps type (H4) [62] with a total current consumption of 10.67 A. These lamps were replaced with LED lamps whose maximum current consumption was 2.4 A. Additionally to this improvement, we increased the DC bus voltage in the lighting system from 12 V to 24 V through a 600 W DC-DC boost converter, searching for a reduction in the current of the LED lamps. These changes allowed a reduction of approximately 72% in the power consumption of the lighting system in comparison with the previous electric system installed in EOLO. In the same way, the initial user interface solely consisted of some signal lights installed on the steering wheel; therefore, an interface with screens was suggested to make the user experience more comfortable.

Some main critical requirements in the project were the reduction of the power consumption in the different nodes created in Figure 2, and the robustness of the CAN-Bus protocol under different events such as voltage transients, malfunction, and inadequate usage or handling by the driver. For this, we selected a 16-bit architecture with the devices (dsPIC33FJ128GP802) [40] that have an Enhanced CAN peripheral (ECAN) module incorporated for CAN-Bus communication [40,63], which have been used in prior projects [15,45,64] with feasible and good results from the educational and technical standpoints. The dsPICs provide advantages such as easiness of programming, tools for compiling and debugging such as MPLABX IDE, XC16, and ICD3, Direct Memory Access (DMA) peripheral in order to transfer data of the ECAN directly to RAM, and a maximum processor speed of 40 Million of Instructions per Second (MIPS). Moreover, the dsPICs have an architecture similar to 8-bit microcontrollers that are used in our curricula in the area of embedded systems, which benefited the learning of the students and accelerated the stage of design, development, and implementation. Some of the results for the systems implemented in the EV will be discussed below. It is worth mentioning that the architecture of the communication was created based on the prior works [65,66,67]. These works describe architectures with a CAN-Bus protocol to monitor variables in hybrid or cargo vehicles.

4.2.2. Front and Rear Lights (Lighting System)

Three LED lamps by side (left and right) were used with a maximum power consumption of 25 W. Each lamp has four power LEDs with its respective LED array for signaling. The intensity of each lamp is controlled by the device FL7760BM6X [43], which is a low-cost linear driver with a hysteresis current controller and a wide input range (8 V to 70 V DC). The FL7760BM6X uses a DC-DC Buck converter topology and it counts with thermal shutdown (TSD) and under-voltage lockout (UVLO) protections. The device receives a PWM signal from dsPIC on the (DIM) pin, which allows it to adjust the duty cycle of the signal to the MOSFET transistor (IRF740AS-Q1) whose operation frequency is 18 KHz. The control operates with the voltage regulation (200 mV $\pm 5\%$) over the resistor $R_{senH}$. Figure 9 shows the implemented schematic in the node. The values of the different components were taken by recommendation of the FL7760BM6X manufacturer [43], and adapted to the requirements of the lighting system according to Figure 2.

Regarding the PWM frequency of 18 KHz for the transistor (Q1), this is configured by the Equation (1).

$$Time{r}_{Reg}=\frac{F_{CY}}{Pre·F_{PWM}}·k$$

(1)

where $F_{CY}=36.85$ MHz, k is the duty cycle $(0\le k\le 1)$, ($pre=1$) is the Timer pre-scale in the PWM peripheral, and $F_{pwm}=18$ KHz. The frequency $F_{CY}$ is obtained directly from the Internal Fast RC oscillator (FRC) of the dsPIC with the phase-locked loop (PLL) feature, providing an accuracy of ±12%. For instance, whether $(k=0.5)$, these values yield to $Time{r}_{Reg}=1024$. Taking into account these elements, each one of the $Time{r}_{Reg}$ values for the front light functions (high, medium, low, and strobe) were characterized as

Table 6. Experimental duty cycle (k) for the front lights. The highest $Time{r}_{Reg}$ value is 2048 for $k=1$.

Light Function	${\mathit{Timer}}_{\mathit{Reg}}$ Value	Duty Cycle (k)
High	1300	0.635
Medium	800	0.4
Low	300	0.147
Strobe	(0–1300)	(0–0.635)

shows.

The different tests made with the lighting system pointed out a critical value for ($k=0.8$) before a thermal failure in (Q1). Therefore, we limited the duty cycle to 0.635 preventing an over-temperature and premature damage in this transistor. The same scheme was followed for the rear lights with the difference that these lights had a current consumption of approximately $I_L=100$ mA, which allowed us to use an array of BJT transistors compatible with automotive transients whose nominal collector current was $I_C=500$ mA instead of the FL7760BM6X device. The appearance of the final PCB for the front lights installed in the vehicle is shown in Figure 10.

4.2.3. Electronic Control Unit (ECU) Node

The ECU node processes and sends different commands for each node in the vehicle. Due to the large number of incoming events for the ECU node, we decided to incorporate a small Real-Time Operating System (RTOS) known as OSA [68]. OSA is a cooperative multitasking RTOS that only occupies around 4% of the dsPIC memory flash whose capacity is 128KB and it allows the designer to select and configure the timer for the different scheduling events, tasks, or semaphores. The tasks for RTOS are configured with their own priority (0–7) where (0) is the highest priority and so on. Two tasks ($t_1$, $t_2$) were set up in the project with a priority of (0) and (2), respectively. The difference between the priorities is because $t_1$ monitors the CAN-Bus communication while $t_2$ is for the handheld control in the steering of the EV.

Task ($t_1$) sends the light intensity and turn signal functions for the nodes that reply with a 29-bit identifier through a polling method with a monitoring period of 2 s. Whether this identifier matches the address allocated in the ECU node, a new command for the node is sent, otherwise, an error is generated. Task ($t_2$) checks the hand-held control and saves the function to be sent to the nodes. The hand-held control is interfaced directly with several General Purpose Input-Output (GPIOs) in the dsPIC with their respective pull-up resistors and RC filters. By the same token, the incoming events from the touch screen are processed by a UART interrupt that saves the information in the ECAN transmission buffer. At the same time, the UART interrupt sends the command to the tachometer screen through another available UART for further processing. In both cases, the UART peripherals have a baud rate of 9600 bps, which reduces data corruption produced by the noisy environment in the electrical system of the vehicle. The RTOS monitors every task ($t_1$, $t_2$) through a time slice of 1 ms provided by the Timer 1 of the dsPIC. A summarized scheme for the ECU functions is depicted in Figure 11.

4.2.4. CAN-Bus Protocol

Each node obtains and processes a command from ECU by a polling method, which means that each node waits for a command in time intervals to prevent collision in the data bus of the CAN communication. A simplified Finite State Machine (FSM) diagram for the front lights node is depicted in Figure 12. A similar FSM diagram of Figure 12 for the front lights was followed for the rear lights with the functions of Figure 2. For readers interesting in the ECAN features of the dsPIC, please check the datasheet [40]. The communication sequence finally developed and implemented with CAN-Bus protocol for each node of the lighting system is described in the following steps:

• When a user starts the system (turn on the EV), state $\left(q_1\right)$, the ECAN peripheral is configured with a default rate of 250 Kbps and a 29-bit identifier (CAN-2.0B) [63]. Additionally, the PWM is started with the mentioned frequency values. The CAN-Bus communication is type broadcast and solely the filters and masks provided in each node can select the adequate message for the node. Although the dsPIC has 15 filters and 3 masks, we decided to use filters (0-1) and masks (0-1) due to the small number of nodes in the communication. Whether no error is detected $(Err=0)$, the FSM passes the control to the state $\left(q_2\right)$. Otherwise, the error is reported to the user $(Err=1)$.
  Notice that also the channels for Interrupt Request (IRQ) for the DMA peripheral must be configured with the appropriate values to manage the ECAN messages. The DMA directly transfers the information to the RAM buffers, which in this case are two, the first one for transmission and the other for reception. The size of each one is [1]*[8] (column vector).
• In the state $\left(q_2\right)$, the node waits for an ECU message which contains the commands for the lights and signals (Mess = 0). In the case of a new message (Mess = 1), depending on the command, the FSM could pass to the states $\left(q_3\right),\left(q_4\right),\left(q_5\right),\left(q_6\right),\left(q_7\right)$ for the light functions (high (LA), medium (LM), low (LB), strobe (Str) and turn off (AP)), respectively. The algorithm remains in its state until another command is received. A similar scheme presented in Figure 12 applies to the turn signals and stationary lights.
• A received command from ECU is a 16-bit concatenate value. That is, a single command contains the functions of the light intensity and the turn signals. An example of this is illustrated in Figure 13. The level of brightness in each level for the front and rear lights is according to the normative in the automotive sector in our country, which agrees with the duty cycle values shown previously in Table 6.
  The previous structure for any command prevents a collision due to the CAN-Bus being used once by the ECU.

  Table 7. Light intensity commands.

  Command	Function
  0	Turn off
  1	High
  2	Medium
  3	Low
  4	Strobe

  and

  Table 8. Turn signal commands.

  Command	Function
  0	Turn off
  1	Turn Right
  2	Turn Left
  3	Stationary lights

  show the commands for light intensity and turn signals according to the structure of Figure 13. The commands can be selected either by the handheld control or the touch screen.

In order to test the CAN-Bus protocol in the different nodes, we followed these steps:

• CAN protocol communication (no functions): The purpose of this test was to check the polling method from the ECU node without functions. The polling method calls every node in Figure 2 and waits for a response. In case of no response, the system generates an alert for the driver. Then, we observed the possible transmission errors. Under this methodology, 30 controlled tests were made and 10 of them (30%) generated an error due to both the reduced time for the polling method and the electrical noisy environment. Our solution was to increase the time of polling in the RTOS and the processing time in each node gradually until the communication was stable. In the same way, we used a shielded cable wire in the vehicle for the CAN-Bus communication.
• CAN protocol communication (with functions): In this test, we checked the functions of the hand-held control and the touch screen with the corrections made in the previous step. A set of functions were sent in short time intervals (10 s) from hand/held control and touch screen, searching for a malfunction in the RTOS or the CAN-Bus protocol. We performed 30 controlled tests without errors.
• Start-up malfunction: In this test, we turned on and off the electrical system several times, searching for errors produced by the voltage transients in the start-up of the nodes. We found that the components more sensible to transients were the screens. Therefore, we designed and implemented a start-up mechanism with a microcontroller and a MOSFET transistor acting as a load switch. The mechanism let the filter bank capacitors in the battery charge for 3 s, and next the electrical system for the nodes was energized. Moreover, the nodes have a start-up period of 2 s.

4.2.5. Driver-focused Human Machine Interface (HMI)

The driver user interface was developed using two screens (Mikroelektronika HMI 7${}^{\prime \prime }$) [42] that count with the features indicated in

Table 9. Technical features of the screens used in the development of the user interface.

Feature	Description
Size	7${}^{\prime \prime }$
Resolution	800 * 480 px
Graphic Controller	FT812/13
Main Microcontroller	FT900Q (32-bit)
Microcontroller’s speed	100 MHz/310 DMIPS
Flash Memory Capacity	256KB
MCU peripherals	CAN (2), SPI (2), I2C (2), I2S (1), UART (2)
Operating Voltage	3.3 V
Manufacturer	Mikroelektronica/Riverdi

.

We selected these screens because they offer high-quality graphics (16.7 M color depth and 400 NIT brightness), low current consumption (around 0.27 A for both screens), and widget support in the compiler toolkit. Regarding the tachometer screen, the speed is updated through a 12 V square signal derived from the electric engine and shifted with optocouplers to the 3.3 V level. The signal changes its frequency ($f{}_t$) from 4.62 Hz up to 30 Hz for a speed between ($\frac{10\mathrm{km}}{\mathrm{h}}$) and ($\frac{100\mathrm{km}}{\mathrm{h}}$). The vehicle’s speed was modeled with a Curtis 1313 controller [69] that allowed it to change the motor’s RPMs from its console taking into account the gear transmission ratio of the EV. This process yields the expression (2):

$$Speed\left(\frac{\mathrm{km}}{\mathrm{h}}\right)=3.5461·f{}_t-6.382$$

(2)

The signal is captured by the input compare module in the dsPIC of the ECU that increases an internal counter/timer relying on the detected rising edges of this signal. When a period of 200 ms has concluded, the counter is checked and converted in frequency to apply Equation (2). Then, the speed value is sent to a tachometer widget.

Each widget or graphical component in the screen has an associated algorithm for the events and a driver. In the driver section, the attributes, I/O pins, and basic shapes that interact with the graphical controller (FT812) are specified. The events section contains the event-driven code when a widget, for example, is touched or clicked and it also has the possibility to interact with General Purpose I/O (GPIOs) and peripherals of the screens. The widgets described in this work were created through the tool Visual TFT [70]. The examples of the tachometer and touch screens are illustrated in Figure 14.

Concerning the touch screen, we designed five sub-screens that have the functions of the main menu, time configuration, light intensity, conditioning air, and CAN-Bus communication test. A Real Time Clock (RTC) with I2C protocol (DS3231) installed on the touch screen allows the time configuration according to the user requirements. When a user changes the light intensity, a UART command is sent to the ECU node for its processing according to Figure 11. The CAN-Bus test command makes a routine to check the communication between nodes, sending the different functions for the vehicle. Both screens contain a security time of 2 s to start up the graphical components and to avoid malfunctioning in them.

4.2.6. Tests of Photovoltaic (PV) Panels

The utility with the described PV panels was to provide electric power to a parallel project concerning the air-conditioning of the EV. The aim of the PV panels is to provide the necessary power to this unit in the vehicle. For that, The PV panels were interfaced to a DC motor of the air-conditioning with three speeds and an operating voltage of 12 V. The DC motor creates an air circulation that cools hot air through a component called evaporator. Air-conditioning is custom and provided by a local manufacturer. Still, this concept is under test and the results exposed are a type of proof-of-concept of the system. We measured the current and voltage of the DC motor + evaporator for the three speeds as

Table 10. Power consumption for the DC motor and evaporator in the air-conditioning unit.

Speed	Voltage (V)	Current (A)	Start-Up Current (A)	Power (W)
1	12.1	2.85	3.38	34.485
2	12.04	4.35	5.18	52.374
3	11.96	6.6	9.1	78.936

shows. The schematic with the connections for the PV panels, 12 V battery, and the evaporator is shown in Figure 15 with a low-cost solar PV panel controller of 20 A. In the same procedure, we identified the start-up current of the motor at these speeds. Unfortunately, the PV panels do not provide this current by themselves and this is the reason because we incorporated a 12 V–5 Ah lead–acid battery. The tests were made as Figure 6e depicts.

Considering these aspects, we performed some tests on a normal day to evaluate the feasibility of the system. The results are condensed in

Table 11. Test of air-conditioning for the three available speeds. BC: Battery current (A), PV: Photovoltaic panel current (A), SV: System voltage (V).

Speed	BC (A)	PV (A)	SV (V)	Time
1	−0.1A	−3.22	12.7	11:45 a.m.
	1.48	−4.72	13	12:15 p.m.
	2.77	−6.12	13.6	12:25 p.m.
	3.31	−6.65	14.4	12:45 p.m.
	2.66	−5.56	13.7	12:55 p.m.
2	−1.61	−3.23	12.5	11:45 a.m.
	−0.08	−4.81	12.7	12:15 p.m.
	1.12	−5.96	13	12:25 p.m.
	2.15	−6.73	13.2	12:45 p.m.
	2.21	−6.73	13.2	12:55 p.m.
3	−3.72	−2.09	12.3	11:45 a.m.
	−2.47	−4.85	12.4	12:15 p.m.
	−1.16	−5.95	12.6	12:25 p.m.
	0.1	−6.77	12.8	12:45 p.m.
	−0.54	−6.6	12.7	12:55 p.m.

. The test was performed at Giradot, a Colombian town located at (74.7799 longitude and 4.3828 latitude). The average solar radiation for the 3 years in this town is 4.72 kWh/m, and the average ambient temperature is 23.06 °C [71]. This town has high rates of irradiance during the year.

In the data of Table 11, there are two elements to notice. The first one is that some values have a negative sign. In this case, this sign indicates that the device (battery or PV panels) is providing current to the air-conditioning. Otherwise, the positive sign indicates that the device (battery) is absorbing the current of the PV panels. The second one is that at certain hours (12:45 p.m.,12:55 p.m.) the maximum currents provided by the PV panels take place, and even in some cases the battery absorbs the current of the PV panels (see speeds 1 and 2). During these hours, the sky was clear and the day was fully sunny. These data point out that sustainable air-conditioning is feasible, and solar panels can generate a maximum of 88.83 W. However, this value must be improved for further developments, especially for speed 3, deeming also aspects such as the temperature of the driver’s cabin, or solar PV panels with better efficiency. These values suppose an approximately 0.6% of increase in the efficiency taking into account the reported consumption of 14,529.9 W of the EV in the tests made in the document [31].

5. Challenges and Recommendations

While the mentioned results show several advantages and success factors in the methodology, some challenges and issues have been detected and they should be taken into mind. A list of these and possible solutions based on the experience in this study are indicated as follows:

• Tutor accompaniment: As described, the tutoring process was difficult in some cases due to the number of tasks and technical requirements in EOLO. Assigning more tutors for the projects and/or defining a smaller but relevant number of activities in each project in coordination with the technical staff in the industry can alleviate in part the issue. Tutor support and accompaniment are key factors in the methodology with PjBL even more when this last is carried out in an industry–academia collaboration.
• Reporting activities and follow-up: Report progress achieved in shorter periods of time, expose advances, challenges, or issues under the peers and tutors, hold more meetings with the tutors, and use ICT tools for collaborative learning can be options explored to improve the report of the students’ activities and their monitoring. Logbooks, short lab reports, and explanatory videos may be considered for the report.
• Administrative and purchasing times: Explore open-source alternatives in hardware and software that could replace the current selection of devices or applications. Inquiry to local distributors for alternative devices or components. Use rapid prototyping techniques with development boards, 3D printers, laser cutters, or CNC machines [72,73,74].
• Sense of community: Create a sense of community with a common goal in the project. All projects are important in the achievement of the objectives defined. The students are not working alone, but their designs, prototypes, or systems must interact with those created by their peers. Foster peer-to-peer collaboration. Create a healthy environment of collaboration and not competition. Share experiences with other colleagues (researchers–professors–faculty staff) even if they are of different knowledge areas. This can aid in obtaining different ideas and standpoints that could nourish the educational and technical objectives of the project.
• SDGs into the curricula: Explore how SDGs and ESD can be incorporated under the lens of the curricula. Identify the competencies (soft and technical) that the students must develop and try to incorporate activities and active learning methodologies that foster SDGs. Clearly define learning objectives based on Bloom’s or Marzano’s taxonomies [75,76], and create, if it is possible, rubrics that help to evaluate the students. Think and create activities that engage the students regarding the SDGs.
• Communication channels: Improve the communication channels between students, tutors, and technical staff involved in the project. Create regular meetings, brainstorming, mind map activities, and expose issues, challenges, or new ideas that will be convenient. Carefully consider students’ views and opinions during the PjBL process.
• Complexity of the projects: The level of complexity is fundamental to demand a high effort in the students, taking into account that this complexity must be attainable by achieving a solution in the projects. The above is based on the importance of reaching achievements and obtaining a solution to the problems proposed, which should have implications in the labor life of the students.
• Further projects: Develop a bank of projects and possible problems in search of their solution, generating improvements to the devices and standardizing the requirements of the projects to be carried out.

6. Conclusions and Further Work

In this article, we described the educational aspects concerning the project EOLO with its implications in the process of learning, motivation, and collaboration of the students. Through an industry–academia collaboration, the students developed a set of projects that gave technical support to the different versions of EOLO. The deployed active learning methodology with PjBL allowed the enhancement of the learning process of the students and encouraged peer collaboration, motivation, and problem-solving skills. These results are aligned with prior works such as [14,18], where students, through discovery and inquiry, searched for solutions for diverse problems in the field of renewable energy. Nonetheless, in counterpart to prior studies which are situated from a holistic view from HEIs [9,13], we think that the initiatives from educators in classrooms are important to achieve and promote SD. With these initiatives, the change in SD can also be fostered in HEIs.

At the educational level, some points such as the tutor’s accompaniment, the project’s complexity, the reduction of the administrative and purchasing times, the follow-up and report of the performed activities, and the sense of community should be revisited to improve the methodology. Additionally, the standpoint of the technical staff of EOLO regarding the performance of the students indicates that the engagement, compromise, curiosity, and cooperation by the students are the most remarkable educational and attitudinal aspects of the methodology. In the same way as the studies [9,13,14,18], we think the most difficult issue in SD is its effective incorporation in the classrooms. Therefore, in this study, we constructed a methodology within a representative project in the renewable energy field. Perhaps, another challenging issue is finding a relevant project that engages students, and that it catalyzes learning, motivation, self-efficacy, etc. This is the main reason why an industry–academia collaboration is suitable to create and develop these types of projects. As aspects to improve, we think that is necessary to widen the scope of the methodology to other engineering and SD fields, reinforcing the previous weakness points found in the prior work [15] and in the current study. Although we believe that the incorporation of 12 students is appropriate considering the features of PjBL, the validation of the methodology with more students could lead us to better improvements and results from the educational standpoint.

At the technical level, this study described a system through a CAN-Bus protocol and a user interface developed with the support of the students that comply with the design requirements and robustness parameters. Moreover, we achieved, with the developed electric and communication architecture, a reduction of 72.3% in the power consumption in the lighting system in contrast with the first electric system deployed in EOLO. Also, some tests regarding the PV panels or the wind turbine suggest that an improvement in the EV’s efficiency is possible. Nonetheless, further development and testing are necessary to fulfill the goal of the efficiency of 11%, and the robustness of the nodes in the EV. Still, several components, nodes, and renewable energy mechanisms are under test in EOLO.

All of these elements allow us to state that the methodology was successful in the objectives and goals proposed. However, the mentioned issues and recommendations should be carefully taken into account for further developments or educational designs of active learning methodologies. Further work will be focused on strengthening the weak points of the methodology and include other projects in the fields of renewable energy, the Internet of Things (IoT), machine learning, and AI in which new students will be involved in EOLO.