Article: Fault-tolerant control of intelligent transportation vehicles based on instant learning and heuristic dynamic planning Journal: International Journal of Reasoning-based Intelligent Systems (IJRIS) 2025 Vol.17 No.8 pp.21 - 28 Abstract: This paper proposes a fault-tolerant control method that integrates real-time learning (JITL) and heuristic dynamic programming (HDP) to address the issues of actuator failures and model uncertainties in intelligent transportation vehicles in dynamic environments. Construct an online fault diagnosis module using a multi-source data-driven framework, and utilise JITL to dynamically update local models to quickly capture system anomalies; design an adaptive controller based on the dual layer optimisation structure of HDP, and compensate for the impact of faults through an evaluation execution network collaborative optimisation strategy. Experimental verification based on the publicly available traffic dataset NGSIM shows that in typical fault scenarios such as sensor failure and actuator offset, the proposed method can significantly improve tracking accuracy and response speed compared to traditional robust control methods, and effectively suppress oscillations caused by interference, verifying the adaptability and reliability of the algorithm in dynamic environments. Inderscience Publishers - linking academia, business and industry through research

Title: Fault-tolerant control of intelligent transportation vehicles based on instant learning and heuristic dynamic planning

Authors: Zhixin Sun

Addresses: Smart Transportation Police Laboratory, Henan Police College, Zhengzhou 450046, China

Abstract: This paper proposes a fault-tolerant control method that integrates real-time learning (JITL) and heuristic dynamic programming (HDP) to address the issues of actuator failures and model uncertainties in intelligent transportation vehicles in dynamic environments. Construct an online fault diagnosis module using a multi-source data-driven framework, and utilise JITL to dynamically update local models to quickly capture system anomalies; design an adaptive controller based on the dual layer optimisation structure of HDP, and compensate for the impact of faults through an evaluation execution network collaborative optimisation strategy. Experimental verification based on the publicly available traffic dataset NGSIM shows that in typical fault scenarios such as sensor failure and actuator offset, the proposed method can significantly improve tracking accuracy and response speed compared to traditional robust control methods, and effectively suppress oscillations caused by interference, verifying the adaptability and reliability of the algorithm in dynamic environments.

Keywords: intelligent transportation; fault-tolerant control; instant learning; heuristic dynamic programming; HDP.

DOI: 10.1504/IJRIS.2025.147145

International Journal of Reasoning-based Intelligent Systems, 2025 Vol.17 No.8, pp.21 - 28

Received: 06 May 2025
Accepted: 24 May 2025
Published online: 10 Jul 2025 *

Title: Fault-tolerant control of intelligent transportation vehicles based on instant learning and heuristic dynamic planning

Keep up-to-date