# Boundary Scenario Generation for HAVs Based on Classification and Local Sampling

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

## 1. Introduction

- (1)
- A novel approach is proposed to obtain a High-Performance Classifier (HPC) for efficiently classifying critical and non-critical scenarios without consuming significant resources. This approach combines the Gaussian Process Classification (GPC) and Support Vector Machine (SVM) algorithms, which mutually detect uncertain scenarios and improve each other’s performance iteratively.
- (2)
- A distance-based method is presented for identifying candidate scenarios using the HPC without executing scenarios. This approach enables the efficient identification of potential boundary scenarios.
- (3)
- This study also uses local sampling iteratively to generate diverse candidate scenarios based on a small set of them. This approach can be applied to generate high-dimensional candidate scenarios efficiently.

## 2. Related Works

## 3. Methodology

#### 3.1. A Distance-Based Approach

#### 3.2. Overall Framework

#### 3.3. Gaussian Processes

#### 3.4. Support Vector Machines

#### 3.5. Obtaining High-Performance Classifiers

#### 3.6. Searching for Candidate Scenarios Based on an HPC

#### 3.7. Generating Diverse Candidate Scenarios by Local Sampling

## 4. Simulation Experiments

#### 4.1. Testing Scenarios

#### 4.2. Intelligent Driver Model

#### 4.3. Stop Conditions

#### 4.4. Distance Thresholds for Determining Candidate or Boundary Scenarios

#### 4.5. Criticality Metric

## 5. Numerical Results and Discussions

#### 5.1. Results and Discussions of Car-Following Scenario Experiments

#### 5.2. Results and Discussions of Cut-In Scenario Experiments

## 6. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Road Traffic Injuries. Available online: https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries (accessed on 1 December 2022).
- Waymo One—Waymo. Available online: https://waymo.com/intl/zh-cn/waymo-one/ (accessed on 26 November 2022).
- Autopilot and Full Self-Driving Capability|Tesla. Available online: https://www.tesla.com/en_AE/support/autopilot-and-full-self-driving-capability (accessed on 26 November 2022).
- Baidu Races Ahead of Tesla With Launch of Robotaxi with Detachable Steering Wheel—WSJ. Available online: https://www.wsj.com/articles/baidu-races-ahead-of-tesla-with-launch-of-robotaxi-with-detachable-steering-wheel-11658396433 (accessed on 26 November 2022).
- Feds Probe Deadly Tesla Crash into Parked Tractor-Trailer at Rest Stop—CBS News. Available online: https://www.cbsnews.com/news/tesla-crash-nhtsa-feds-probe-deadly-florida-rest-stop/ (accessed on 26 November 2022).
- Artificial Intelligence & Autopilot|Tesla. Available online: https://www.tesla.com/AI (accessed on 26 November 2022).
- Cai, J.; Deng, W.; Guang, H.; Wang, Y.; Li, J.; Ding, J. A Survey on Data-Driven Scenario Generation for Automated Vehicle Testing. Machines
**2022**, 10, 1101. [Google Scholar] [CrossRef] - Kalra, N.; Paddock, S.M. Driving to Safety: How Many Miles of Driving Would It Take to Demonstrate Autonomous Vehicle Reliability? Transp. Res. Part A Policy Pract.
**2016**, 94, 182–193. [Google Scholar] [CrossRef] - Ulbrich, S.; Menzel, T.; Reschka, A.; Schuldt, F.; Maurer, M. Defining and Substantiating the Terms Scene, Situation, and Scenario for Automated Driving. In Proceedings of the 2015 IEEE 18th International Conference on Intelligent Transportation systems, Gran Canaria, Spain, 15–18 September 2015; pp. 982–988. [Google Scholar]
- Menzel, T.; Bagschik, G.; Maurer, M. Scenarios for Development, Test and Validation of Automated Vehicles. In Proceedings of the 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 26–30 June 2018; pp. 1821–1827. [Google Scholar]
- Bagschik, G.; Menzel, T.; Maurer, M. Ontology Based Scene Creation for the Development of Automated Vehicles. In Proceedings of the 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 26–30 June 2018; pp. 1813–1820. [Google Scholar]
- Scholtes, M.; Westhofen, L.; Turner, L.R.; Lotto, K.; Schuldes, M.; Weber, H.; Wagener, N.; Neurohr, C.; Bollmann, M.H.; Körtke, F. 6-Layer Model for a Structured Description and Categorization of Urban Traffic and Environment. IEEE Access
**2021**, 9, 59131–59147. [Google Scholar] [CrossRef] - Roesener, C.; Sauerbier, J.; Zlocki, A.; Fahrenkrog, F.; Wang, L.; Várhelyi, A.; de Gelder, E.; Dufils, J.; Breunig, S.; Mejuto, P. A Comprehensive Evaluation Approach for Highly Automated Driving. In Proceedings of the 25th International Technical Conference on the Enhanced Safety of Vehicles (ESV) National Highway Traffic Safety Administration, Detroit, MI, USA, 5–8 June 2017. [Google Scholar]
- Song, Q.; Tan, K.; Runeson, P.; Persson, S. Critical Scenario Identification for Realistic Testing of Autonomous Driving Systems. Softw. Qual. J. 2022. Available online: https://assets.researchsquare.com/fifiles/rs-1280095/v1_covered.pdf?c=1642707391 (accessed on 10 February 2023).
- Mullins, G.E.; Stankiewicz, P.G.; Hawthorne, R.C.; Gupta, S.K. Adaptive Generation of Challenging Scenarios for Testing and Evaluation of Autonomous Vehicles. J. Syst. Softw.
**2018**, 137, 197–215. [Google Scholar] [CrossRef] - Wang, X.; Zhang, S.; Peng, H. Comprehensive Safety Evaluation of Highly Automated Vehicles at the Roundabout Scenario. IEEE Trans. Intell. Transp. Syst.
**2022**, 23, 20873–20888. [Google Scholar] [CrossRef] - Thorn, E.; Kimmel, S.C.; Chaka, M.; Hamilton, B.A. A Framework for Automated Driving System Testable Cases and Scenarios; Department of Transportation, National Highway Traffic Safety: Washington, DC, USA, 2018.
- Batsch, F.; Daneshkhah, A.; Cheah, M.; Kanarachos, S.; Baxendale, A. Performance Boundary Identification for the Evaluation of Automated Vehicles Using Gaussian Process Classification. In Proceedings of the 2019 IEEE Intelligent Transportation Systems Conference (ITSC), Auckland, New Zealand, 27–30 October 2019; pp. 419–424. [Google Scholar]
- Birkemeyer, L.; Pett, T.; Vogelsang, A.; Seidl, C.; Schaefer, I. Feature-Interaction Sampling for Scenario-Based Testing of Advanced Driver Assistance Systems. In Proceedings of the 16th International Working Conference on Variability Modelling of Software-Intensive Systems, Florence, Italy, 23–25 February 2022; pp. 1–10. [Google Scholar]
- Xia, Q.; Duan, J.; Gao, F.; Chen, T.; Yang, C. Automatic Generation Method of Test Scenario for Adas Based on Complexity. In Proceedings of the Intelligent and Connected Vehicles Symposium, Los Angeles, CA, USA, 11–14 June 2017; Volume 1. [Google Scholar]
- Zhao, D. Accelerated Evaluation of Automated Vehicles. Ph.D. Thesis, The University of Michigan, Ann Arbor, MI, USA, 2016. [Google Scholar]
- Huang, Z.; Lam, H.; LeBlanc, D.J.; Zhao, D. Accelerated Evaluation of Automated Vehicles Using Piecewise Mixture Models. IEEE Trans. Intell. Transp. Syst.
**2017**, 19, 2845–2855. [Google Scholar] [CrossRef][Green Version] - Zhang, S.; Peng, H.; Zhao, D.; Tseng, H.E. Accelerated Evaluation of Autonomous Vehicles in the Lane Change Scenario Based on Subset Simulation Technique. In Proceedings of the 2018 21st International Conference on Intelligent Transportation Systems (ITSC), Maui, HI, USA, 4–7 November 2018; pp. 3935–3940. [Google Scholar]
- Xu, Y.; Zou, Y.; Sun, J. Accelerated Testing for Automated Vehicles Safety Evaluation in Cut-in Scenarios Based on Importance Sampling, Genetic Algorithm and Simulation Applications. J. Intell. Connect. Veh.
**2018**, 1, 28–38. [Google Scholar] [CrossRef][Green Version] - Li, G.; Li, Y.; Jha, S.; Tsai, T.; Sullivan, M.; Hari, S.K.S.; Kalbarczyk, Z.; Iyer, R. AV-FUZZER: Finding Safety Violations in Autonomous Driving Systems. In Proceedings of the 2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE), Coimbra, Portugal, 12–15 October 2020; pp. 25–36. [Google Scholar]
- Klischat, M.; Liu, E.I.; Holtke, F.; Althoff, M. Scenario Factory: Creating Safety-Critical Traffic Scenarios for Automated Vehicles. In Proceedings of the 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Rhodes, Greece, 20–23 September 2020; pp. 1–7. [Google Scholar]
- Zhu, B.; Zhang, P.; Zhao, J.; Deng, W. Hazardous Scenario Enhanced Generation for Automated Vehicle Testing Based on Optimization Searching Method. IEEE Trans. Intell. Transp. Syst.
**2021**, 23, 7321–7331. [Google Scholar] [CrossRef] - Feng, S.; Feng, Y.; Sun, H.; Zhang, Y.; Liu, H.X. Testing Scenario Library Generation for Connected and Automated Vehicles: An Adaptive Framework. IEEE Trans. Intell. Transp. Syst.
**2020**, 23, 1213–1222. [Google Scholar] [CrossRef] - Tuncali, C.E.; Fainekos, G. Rapidly-Exploring Random Trees-Based Test Generation for Autonomous Vehicles. arXiv
**2019**, arXiv:1903.10629. [Google Scholar] - Karunakaran, D.; Worrall, S.; Nebot, E. Efficient Falsification Approach for Autonomous Vehicle Validation Using a Parameter Optimisation Technique Based on Reinforcement Learning. arXiv
**2020**, arXiv:2011.07699. [Google Scholar] - Baumann, D.; Pfeffer, R.; Sax, E. Automatic Generation of Critical Test Cases for the Development of Highly Automated Driving Functions. In Proceedings of the 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring), Helsinki, Finland, 25–28 April 2021; pp. 1–5. [Google Scholar]
- Xu, L.; Zhang, C.; Liu, Y.; Wang, L.; Li, L. Worst Perception Scenario Search for Autonomous Driving. In Proceedings of the 2020 IEEE Intelligent Vehicles Symposium (IV), Las Vegas, NV, USA, 19 October–13 November 2020; pp. 1702–1707. [Google Scholar]
- Koren, M.; Alsaif, S.; Lee, R.; Kochenderfer, M.J. Adaptive Stress Testing for Autonomous Vehicles. In Proceedings of the 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 26–30 June 2018; pp. 1–7. [Google Scholar]
- Koren, M.; Nassar, A.; Kochenderfer, M.J. Finding Failures in High-Fidelity Simulation Using Adaptive Stress Testing and the Backward Algorithm. In Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 27 September–1 October 2021; pp. 5944–5949. [Google Scholar]
- Lagaros, N.D.; Plevris, V.; Kallioras, N.A. The Mosaic of Metaheuristic Algorithms in Structural Optimization. Arch. Comput. Methods Eng.
**2022**, 29, 5457–5492. [Google Scholar] [CrossRef] - Li, B.; Wu, G.; He, Y.; Fan, M.; Pedrycz, W. An Overview and Experimental Study of Learning-Based Optimization Algorithms for the Vehicle Routing Problem. IEEE/CAA J. Autom. Sin.
**2022**, 9, 1115–1138. [Google Scholar] [CrossRef] - Bussler, A.; Hartjen, L.; Philipp, R.; Schuldt, F. Application of Evolutionary Algorithms and Criticality Metrics for the Verification and Validation of Automated Driving Systems at Urban Intersections. In Proceedings of the 2020 IEEE Intelligent Vehicles Symposium (IV), Las Vegas, NV, USA, 19 October–13 November 2020; pp. 128–135. [Google Scholar]
- van der Horst, R.; Hogema, J. Time-to-Collision and Collision Avoidance Systems. Sci. Eng. Med. 1993. In 6th ictct workshop (pp. 1–12). Salzburg: ICTCT. Available online: https://www.academia.edu/download/35311069/Horst.pdf (accessed on 10 February 2023).
- Eggert, J. Predictive Risk Estimation for Intelligent ADAS Functions. In Proceedings of the 17th International IEEE Conference on Intelligent Transportation Systems (ITSC), Qingdao, China, 8–11 October 2014; pp. 711–718. [Google Scholar]
- Minderhoud, M.M.; Bovy, P.H.L. Extended Time-to-Collision Measures for Road Traffic Safety Assessment. Accid. Anal. Prev.
**2001**, 33, 89–97. [Google Scholar] [CrossRef] [PubMed] - Allen, B.L.; Shin, B.T.; Cooper, P.J. Analysis of Traffic Conflicts and Collisions; Transportation Research Record Department of Civil Engineering; McMaster University: Hamilton, ON, Canada, 1978. [Google Scholar]
- Varhelyi, A. Drivers’ Speed Behaviour at a Zebra Crossing: A Case Study. Accid. Anal. Prev.
**1998**, 30, 731–743. [Google Scholar] [CrossRef] - Feng, S.; Feng, Y.; Yu, C.; Zhang, Y.; Liu, H.X. Testing Scenario Library Generation for Connected and Automated Vehicles, Part I: Methodology. IEEE Trans. Intell. Transp. Syst.
**2020**, 22, 1573–1582. [Google Scholar] [CrossRef][Green Version] - Huber, B.; Herzog, S.; Sippl, C.; German, R.; Djanatliev, A. Evaluation of Virtual Traffic Situations for Testing Automated Driving Functions Based on Multidimensional Criticality Analysis. In Proceedings of the 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Rhodes, Greece, 20–23 September 2020; pp. 1–7. [Google Scholar]
- Hauer, F.; Pretschner, A.; Holzmüller, B. Fitness Functions for Testing Automated and Autonomous Driving Systems. In Proceedings of the International Conference on Computer Safety, Reliability, and Security, Indianapolis, IN, USA, 19–22 September 2021; pp. 69–84. [Google Scholar]
- Brunke, L.; Greeff, M.; Hall, A.W.; Yuan, Z.; Zhou, S.; Panerati, J.; Schoellig, A.P. Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning. Annu. Rev. Control Robot. Auton. Syst.
**2022**, 5, 411–444. [Google Scholar] [CrossRef] - Zhu, Y.; Du, Y.; Wang, Y.; Xu, Y.; Zhang, J.; Liu, Q.; Wu, S. A Survey on Deep Graph Generation: Methods and Applications. arXiv
**2022**, arXiv:2203.06714. [Google Scholar] - ben Abdessalem, R.; Nejati, S.; Briand, L.C.; Stifter, T. Testing Vision-Based Control Systems Using Learnable Evolutionary Algorithms. In Proceedings of the 2018 IEEE/ACM 40th International Conference on Software Engineering (ICSE), Gothenburg, Sweden, 27 May–3 June 2018; pp. 1016–1026. [Google Scholar]
- Wang, Y.; Yu, R.; Qiu, S.; Sun, J.; Farah, H. Safety Performance Boundary Identification of Highly Automated Vehicles: A Surrogate Model-Based Gradient Descent Searching Approach. IEEE Trans. Intell. Transp. Syst.
**2022**, 23, 23809–23820. [Google Scholar] [CrossRef] - Batsch, F.; Daneshkhah, A.; Palade, V.; Cheah, M. Scenario Optimisation and Sensitivity Analysis for Safe Automated Driving Using Gaussian Processes. Appl. Sci.
**2021**, 11, 775. [Google Scholar] [CrossRef] - Liu, H.; Ong, Y.-S.; Shen, X.; Cai, J. When Gaussian Process Meets Big Data: A Review of Scalable GPs. IEEE Trans. Neural. Netw. Learn. Syst.
**2020**, 31, 4405–4423. [Google Scholar] [CrossRef][Green Version] - Cervantes, J.; Garcia-Lamont, F.; Rodríguez-Mazahua, L.; Lopez, A. A Comprehensive Survey on Support Vector Machine Classification: Applications, Challenges and Trends. Neurocomputing
**2020**, 408, 189–215. [Google Scholar] [CrossRef] - Karamizadeh, S.; Abdullah, S.M.; Halimi, M.; Shayan, J.; javad Rajabi, M. Advantage and Drawback of Support Vector Machine Functionality. In Proceedings of the 2014 international conference on computer, communications, and control technology (I4CT), Langkawi, Malaysia, 2–4 September 2014; pp. 63–65. [Google Scholar]
- Rasmussen, C.E. Gaussian Processes in Machine Learning. In Summer School on Machine Learning; Springer: Berlin/Heidelberg, Germany, 2003; pp. 63–71. [Google Scholar]
- Schulz, E.; Speekenbrink, M.; Krause, A. A Tutorial on Gaussian Process Regression: Modelling, Exploring, and Exploiting Functions. J. Math. Psychol.
**2018**, 85, 1–16. [Google Scholar] [CrossRef] - Seeger, M. Gaussian Processes for Machine Learning. Int. J. Neural. Syst.
**2004**, 14, 69–106. [Google Scholar] [CrossRef][Green Version] - Melo, J. Gaussian Processes for Regression: A Tutorial. Technical. Report
**2012**. Available online: http://paginas.fe.up.pt/~dee10008/papers/201201_report_ML_jmelo.pdf (accessed on 10 February 2023). - Cherkassky, V.; Mulier, F.M. Learning from Data: Concepts, Theory, and Methods; John Wiley & Sons: Hoboken, NJ, USA, 2007; ISBN 0470140518. [Google Scholar]
- Burges, C.J.C. A Tutorial on Support Vector Machines for Pattern Recognition. Data Min. Knowl. Discov.
**1998**, 2, 121–167. [Google Scholar] [CrossRef] - Somvanshi, M.; Chavan, P.; Tambade, S.; Shinde, S.V. A Review of Machine Learning Techniques Using Decision Tree and Support Vector Machine. In Proceedings of the 2016 international conference on computing communication control and automation (ICCUBEA), Pune, India, 12–13 August 2016; pp. 1–7. [Google Scholar]
- Niranjan, M. Support Vector Machines: A Tutorial Overview and Critical Appraisal. In Proceedings of the IEE Colloquium on Applied Statistical Pattern Recognition (Ref. No. 1999/063), Brimingham, UK, 20 April 1999; pp. 1–2. [Google Scholar]
- Mammone, A.; Turchi, M.; Cristianini, N. Support Vector Machines. Wiley Interdiscip. Rev. Comput. Stat.
**2009**, 1, 283–289. [Google Scholar] [CrossRef] - Maldonado-Bascón, S.; Lafuente-Arroyo, S.; Gil-Jimenez, P.; Gómez-Moreno, H.; López-Ferreras, F. Road-Sign Detection and Recognition Based on Support Vector Machines. IEEE Trans. Intell. Transp. Syst.
**2007**, 8, 264–278. [Google Scholar] [CrossRef][Green Version] - Zhang, H.; Zhou, H.; Sun, J.; Tian, Y. Risk Assessment of Highly Automated Vehicles with Naturalistic Driving Data: A Surrogate-Based Optimization Method. In Proceedings of the 2022 IEEE Intelligent Vehicles Symposium (IV), Aachen, Germany, 4–9 June 2022; pp. 580–585. [Google Scholar]

**Figure 1.**The given scenario and its adjacent, adverse scenarios and the nearest adverse scenario. Blue circles are adjacent scenarios; Green circles are adverse scenarios; The pink circle is the nearest adverse scenario to the given scenario. The red circle is the given scenario.

**Figure 2.**Sketch of soft classification and nonlinear classification based on SVM. (

**a**) Separation hyperplanes; (

**b**) Optimal hyperplanes. Blue and red dots are data belonging to different classes.

**Figure 6.**Flow chart of deriving more candidate scenarios based on a small set of candidate scenarios.

**Figure 7.**Sketches of the car-following scenario and the cut-in scenario. (

**a**) The car-following scenario; (

**b**) The cut-in scenario.

**Figure 8.**The performance of the four classifiers in classifying critical and non-critical car-following scenarios.

**Figure 10.**Size curves of training datasets for GSVM and GGPC in experiments involving car-following scenarios. The blue and red curves belong to GGPC and GSVM, respectively.

**Figure 11.**Candidate scenarios detected based on GGPC and their fitting surface. Red circles are critical scenarios, while blue circles are non-critical scenarios. In addition, the semi-transparent blue surface is the performance boundary.

**Figure 12.**The performance boundary and test samples. The blue surface is the performance boundary between critical and non-critical scenarios. Red circles are critical scenarios, while green circles are non-critical scenarios. To enhance the visualization, we only display 2% of non-critical and 20% of critical scenarios from the test dataset.

**Figure 13.**The performance boundary and the candidate scenarios on it. A total of 99.87% of them are found to be boundary scenarios. The mean distance between boundary scenarios and their nearest adverse scenarios is 0.0057, much smaller than the threshold of 0.02.

**Figure 14.**Size curves of the training datasets for GSVM and GGPC in experiments involving cut-in scenarios. The blue and red curves belong to GSVM and GGPC, respectively.

**Figure 16.**Derived, initial, and lonely candidate scenarios. Red circles are initial candidate scenarios. Green and red circles are derived from candidate and lonely scenarios, respectively. (

**a**) Iteration = 1; (

**b**) Iteration = 3; (

**c**) Iteration = 10; (

**d**) Iteration = 44.

**Figure 17.**The distribution of distances between candidate scenarios and their nearest adverse scenarios.

Name | Equation |
---|---|

Linear Kernel | $k\left(x,y\right)={x}^{\mathrm{T}}y+c$ |

Polynomial Kernel | $k\left(x,y\right)=a{x}^{\mathrm{T}}y+{c}^{d}$ |

Radial Basis Function | $k\left(x,y\right)=\mathrm{exp}(-\gamma {||x-y||}^{2})$ |

Gaussian Kernel | $k\left(x,y\right)=\mathrm{exp}(-\frac{{||x-y||}^{2}}{2{\sigma}^{2}})$ |

Exponential Kernel | $k\left(x,y\right)=\mathrm{exp}(-\frac{||x-y||}{2\sigma})$ |

Description | Key Parameter | Value Range |
---|---|---|

the initial distance between the two vehicles | ${S}_{x,0}$ | (15, 100) m |

the initial velocity of the ego vehicle | ${v}_{\mathrm{x},0}^{\mathrm{ego}}$ | (5, 40) m/s |

the initial velocity of the reference vehicle | ${v}_{\mathrm{x},0}^{\mathrm{ref}}$ | (5, 40) m/s |

Description | Key Parameter | Value Range |
---|---|---|

the initial longitudinal distance between two vehicles | ${S}_{x,0}$ | (15, 100 m |

initial lateral distance between two vehicles | ${S}_{y,0}$ | (1.9, 3.8) m |

initial longitudinal speed of the ego vehicle | ${v}_{x,0}^{\mathrm{ego}}$ | (10, 40) m/s |

initial lateral speed of the reference vehicle | ${v}_{y,0}^{\mathrm{ref}}$ | (0.5, 1.75) m/s |

initial longitudinal speed of the reference vehicle | ${v}_{x,0}^{\mathrm{ref}}$ | (10, 35) m/s |

Description | Key Parameter | Value |
---|---|---|

vehicle width | ${w}_{veh}$ | 1.8 m |

vehicle length | ${l}_{veh}$ | 5 m |

lane width | ${w}_{lane}$ | 3.8 m |

the maximum duration of the scenario | ${T}_{\mathrm{max}}$ | 10 m |

Description | Key Parameter | Typical Value |
---|---|---|

Desired velocity | ${v}_{d}$ | 29.8 m/s |

Safe time headway | $T$ | 1.6 s |

Maximum acceleration | $a$ | 2.62 m/s^{2} |

Desired deceleration | $b$ | 2.67 m/s^{2} |

Acceleration exponent | $\delta $ | 4 |

Jam distance 1 | ${s}_{0}$ | 1 m |

Jam distance 2 | ${s}_{1}$ | 2 m |

Vehicle length | $l$ | 5 m |

Metrics | GPC | SVM | GGPC | GSVM |
---|---|---|---|---|

True Positive Rate (TPR) | 66.02% | 82.90% | 99.66% | 98.75% |

True Negative Rate (TNR) | 93.86% | 100% | 99.87% | 98.77% |

False Positive Rate (FPR) | 6.14% | 0 | 0.13% | 1.23% |

False Negative Rate (FNR) | 33.98% | 17.10% | 0.34% | 1.25% |

Overall Accuracy | 91.42% | 98.50% | 99.85% | 98.77% |

Method | Number of Candidate Scenarios | Number of Boundary Scenarios | The Proportion of Boundary Scenarios | Mean Distance to the NAS |
---|---|---|---|---|

GPC | 7866 | 1346 | 17.11% | 0.017 |

SVM | 2120 | 1308 | 61.70% | 0.017 |

GGPC | 2329 | 2301 | 98.80% | 0.015 |

GSVM | 2348 | 1562 | 66.52% | 0.017 |

GPC | SVM | GGPC | GSVM | |
---|---|---|---|---|

True Positive Rate | 60.65% | 49.52% | 97.10% | 97.90% |

True Negative Rate | 88.11% | 100% | 99.47% | 99.45% |

False Positive Rate | 11.89% | 0 | 0.53% | 0.55% |

False Negative Rate | 39.35% | 50.48% | 2.90% | 2.10% |

Overall Right Rate | 86.43% | 96.91% | 99.33% | 99.36% |

Method | Candidate Scenarios | Boundary Scenarios | Proportion of Boundary Scenarios | Mean Distance to the NAS |
---|---|---|---|---|

GPC | 489 | 50 | 10.22% | 0.033 |

SVM | 52 | 17 | 32.69% | 0.037 |

GGPC | 111 | 88 | 79.28% | 0.031 |

GSVM | 106 | 97 | 91.51% | 0.031 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Cai, J.; Deng, W.; Wang, Y.; Guang, H.; Li, J.; Ding, J. Boundary Scenario Generation for HAVs Based on Classification and Local Sampling. *Machines* **2023**, *11*, 426.
https://doi.org/10.3390/machines11040426

**AMA Style**

Cai J, Deng W, Wang Y, Guang H, Li J, Ding J. Boundary Scenario Generation for HAVs Based on Classification and Local Sampling. *Machines*. 2023; 11(4):426.
https://doi.org/10.3390/machines11040426

**Chicago/Turabian Style**

Cai, Jinkang, Weiwen Deng, Ying Wang, Haoran Guang, Jiangkun Li, and Juan Ding. 2023. "Boundary Scenario Generation for HAVs Based on Classification and Local Sampling" *Machines* 11, no. 4: 426.
https://doi.org/10.3390/machines11040426