Abstract:The path tracking control system is a crucial component for autonomous vehicles; it is challenging to realize accurate tracking control when approaching a wide range of uncertain situations and dynamic environments, particularly when such control must perform as well as, or better than, human drivers. While many methods provide state-of-the-art tracking performance, they tend to emphasize constant PID control parameters, calibrated by human experience, to improve tracking accuracy. A detailed analysis shows th… Show more
“…To solve the problem of feature representation and online learning capability in learning control of uncertain dynamic systems, a multicore online RL method for path tracking control was proposed in the literature [24], where a multicore feature learning framework was designed based on pairwise heuristic planning, and simulations under S-curve and urban road conditions verified that the controller has better tracking accuracy and stability performance than LQR controller and PP controller.Ma et al combined RL and PID were combined to propose a self-seeking optimal path tracking control based on the interactive learning mechanism of the RL framework to achieve online optimization of the PID control parameters, and the simulation and real vehicle tests proved that the control method has better tracking performance in high-speed conditions (maximum speed above 100 km/h) [104].…”
Driverless technology aims to improve driving safety, accuracy and
comfort. Path tracking is a basic component of the motion control module
of autonomous vehicles, and its control algorithm directly affects the
path tracking effect. Based on the preliminary results of the
application of path tracking control algorithm, this paper analyzes the
principles, advantages and disadvantages, applications and current
research progress of the path tracking algorithm under different working
conditions from the perspective of different working conditions at low
speed and high speed, and provides an outlook on the future development,
aiming to provide reference for future in-depth research.
“…To solve the problem of feature representation and online learning capability in learning control of uncertain dynamic systems, a multicore online RL method for path tracking control was proposed in the literature [24], where a multicore feature learning framework was designed based on pairwise heuristic planning, and simulations under S-curve and urban road conditions verified that the controller has better tracking accuracy and stability performance than LQR controller and PP controller.Ma et al combined RL and PID were combined to propose a self-seeking optimal path tracking control based on the interactive learning mechanism of the RL framework to achieve online optimization of the PID control parameters, and the simulation and real vehicle tests proved that the control method has better tracking performance in high-speed conditions (maximum speed above 100 km/h) [104].…”
Driverless technology aims to improve driving safety, accuracy and
comfort. Path tracking is a basic component of the motion control module
of autonomous vehicles, and its control algorithm directly affects the
path tracking effect. Based on the preliminary results of the
application of path tracking control algorithm, this paper analyzes the
principles, advantages and disadvantages, applications and current
research progress of the path tracking algorithm under different working
conditions from the perspective of different working conditions at low
speed and high speed, and provides an outlook on the future development,
aiming to provide reference for future in-depth research.
“…Control algorithms, as the core component of path tracking control systems, have been the focus of most researchers, who aim to enhance the precision of path tracking control and its robustness. Currently, widely used control algorithms include PID control [7][8][9], fuzzy control [10][11][12], model predictive control [13][14][15], and sliding mode control [16][17][18]. In [8], the authors proposed a steering method that integrates reinforcement learning with traditional PID controllers.…”
Section: Introductionmentioning
confidence: 99%
“…Currently, widely used control algorithms include PID control [7][8][9], fuzzy control [10][11][12], model predictive control [13][14][15], and sliding mode control [16][17][18]. In [8], the authors proposed a steering method that integrates reinforcement learning with traditional PID controllers. This approach employs an RL framework with interactive learning mechanisms, enabling adaptive adjustment of the PID control parameters and maintaining excellent tracking accuracy even on complex trajectories.…”
This paper proposes an intelligent vehicle lateral-vertical cooperative control method based on optimized preview distance to address the coupling and conflict issues between the path tracking system and the handling stability in the process of controlling the motion of intelligent vehicles. To begin with, an intelligent vehicle path tracking control system based on expected yaw velocity has been designed by establishing a three-degree-of-freedom dynamic model for intelligent vehicles. Then, we analyzed the mechanism by which changes in vehicle speed, road curvature, and preview distance affect the accuracy of vehicle path tracking and handling stability. Considering the "human-vehicle-road" system in intelligent transportation systems, critical values for collision and instability were set. Furthermore, we designed a proactive optimization method for preview distance under different working conditions, using an optimization algorithm to improve path tracking accuracy while ensuring vehicle stability, based on the lateral displacement deviation and lateral lateral orientation deviation representing the accuracy of path tracking, as well as the lateral acceleration representing handling stability. Finally, HIL platform test was conducted. The simulation and test results show that the optimized path tracking algorithm reduces lateral deviation to as low as 0.05 meters, and the stability constraint control in the algorithm can be triggered promptly even under extreme conditions. The research findings provide a theoretical reference for the lateral and vertical coordinated control of intelligent vehicles in the future.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.