|
An Optimization and Validation Method to Detect the Collision Scenarios and Identifying the Safety Specification of Highly Automated Driving Vehicle |
|
Marzana Khatun(IFM – Institut für Fahrerassistenz und vernetzte Mobilität, Germany) |
| Modeling and simulation techniques are a necessity to solve the problems and aid the automated driving verification and validation process. In terms of functional safety and safety of the intended functionality the number of hazardous scenarios increases that need to be reduced. Therefore, this paper proposes a probability approach like the Monte Carlo method at the logical scenario level. Furthermore, the result realized by the Monte Carlo experiment has been used to model the concrete scenarios in CarMaker in a time-efficient manner and scenario database can be created for further research. |
|
|
Frequency Modulation about Electrical Noise Effect in Radar Application for Autonomous Vehicle Systems |
|
SungHoon Kim(Infineon Technologies, Korea), JongGyu Park(Mando corporation, Korea) |
| It has become an important trend to implement Autonomous vehicle systems. Many of OEM’s are studying and developing Autonomous system at their own vehicle. For Autonomous Vehicle system, many sensor solutions are considered and implemented. Radar application is the most important Sensor solution for Autonomous vehicle systems. In Radar application case, it has robustness in weather condition but electrical noise effect cause very critical issue for object detection and target classification in Radar system. In this paper, it suggests frequency modulation method by hardware peripheral of microcontroller and PMIC about electrical noise effect in radar application. |
|
|
Lane Change Intention Inference of Surrounding Vehicle: Comparative Study on Relevance Vector Machine (RVM) and Support Vector Machine (SVM) |
|
Jin Ho Yang, Daejung Kim, Chung Choo Chung(Hanyang University, Korea) |
| This paper presents a methodology to infer the intention to lane change of Surrounding Vehicle (SV) by designing a classifier using a Relevance Vector Machine (RVM). Estimating the intentions precisely of SV is one of the key technologies in autonomous driving. In particular, the lane change of SV is a situation that can be frequently observed while driving, and the behavior of the host vehicle may be affected by the SVs. Statistically, the proposed RVM-based method succeeded in predicting the lane change of the surrounding vehicle faster than both the RADAR sensor and the SVM. |
|
|
Energy-Optimal Speed Planning for Connected and Autonomous Electric Vehicles |
|
Hoonhee Kim, Jinwoo Bae, Sunwoo Kim, Kwangki Kim(Inha University, Korea) |
| This paper considers energy-optimal speed planning for connected and automated electric vehicles. For optimal speed planning, we consider three driving scenarios depending on driving environments of the upcoming traffic lights, road slope information, and safety distance gap between the host vehicle and the preceding vehicle. For each scenarios, different optimal control problems are formulated with consideration of multi-objectives and powertrain limits as well as vehicle position, speed, acceleration, and jerk constraints. To find optimal solutions, methods of dynamic programming are applied. The proposed optimal control problems and solutions are illustrated with simulation case studies. |
|
|
Barrier Lyapunov Function-Based Safe Reinforcement Learning Algorithm for Autonomous Vehicles with System Uncertainty |
|
Yuxiang Zhang, Xiaoling Liang, Shuzhi Sam Ge(National University of Singapore, Singapore), Bingzhao Gao(Tongji University, China), Tong Heng Lee(National University of Singapore, Singapore) |
| Guaranteed safety and performance remain technically critical and practically challenging for the wide deployment of autonomous vehicles. For such safety-critical systems, it will certainly be a requirement that safe performance should be ensured even during the reinforcement learning period in the presence of system uncertainty. To address this issue, a Barrier Lyapunov Function-based safe reinforcement learning algorithm (BLF-SRL) is proposed here for the formulated nonlinear system in strict-feedback form. This approach appropriately arranges the Barrier Lyapunov Function item into the optimized backstepping control method to constrain the state-variables in the designed safety region. |
| |