Title: A novel obstacle avoidance strategy for autonomous vehicle based on dynamic safety boundary

Authors: Wenbo Li; Han Jin; Jun-Guo Lu; Hongyi Kang; Qing-Hao Zhang

Addresses: Institute of Micro-Nano Science and Technology, National Key Laboratory of Advanced Micro and Nano Manufacture Technology, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai – 200240, China ' Institute of Micro-Nano Science and Technology, National Key Laboratory of Advanced Micro and Nano Manufacture Technology, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai – 200240, China; Medical School, Henan University, Kaifeng – 475004, Henan Province, China; National Engineering Research Center for Nanotechnology, Shanghai – 200241, China; Wuzhen Laboratory, Tongxiang – 314500, Zhejiang Province, China ' Department of Automation, Shanghai Jiao Tong University, Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai 200240, China ' Department of Automation, Shanghai Jiao Tong University, Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai 200240, China ' Department of Automation, Shanghai Jiao Tong University, Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai 200240, China

Abstract: Obstacle avoidance is a pivotal function in autonomous driving to ensure vehicle safety and operational efficiency, particularly in complex dynamic scenarios. However, traditional methods often fall short of effectively addressing the variability and high-risk conditions inherent in real-world environments, thereby limiting their practical applicability. To address these limitations, this paper proposes a novel obstacle avoidance strategy based on a dynamic safety boundary. Specifically, candidate paths are first generated through local path planning and subsequently evaluated for safety using a collision detection algorithm. The static safety boundary is then extended into a dynamic model, enabling real-time adjustments to safety distances and facilitating precise obstacle avoidance. Experimental results across multiple real-world scenarios demonstrate that the proposed approach significantly improves obstacle avoidance performance with only a marginal increase in response time in complex dynamic scenarios, thus highlighting its superior reliability and practical applicability.

Keywords: obstacle avoidance; autonomous driving; safety boundary; local path planning; collision detection algorithm; complex dynamic scenario.

DOI: 10.1504/IJSCC.2026.155209

International Journal of Systems, Control and Communications, 2026 Vol.17 No.3, pp.301 - 326

Received: 02 Feb 2025
Accepted: 12 May 2025
Published online: 29 Jul 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article