Stanley Controller based Autonomous Path planning and Tracking in Self-Driving Cars

Abstract

Autonomous systems have the ability to replace human-performed tasks like personal assistants in residential or commercial settings. Self-driving cars, which have shown potential, are one area of significant interest in AI. This may include anticipating the activities and goals of people, such as pedestrians, as well as those of other vehicles. The creation of high-resolution images and sophisticated obstacle-clearing manoeuvres at high speeds are other developments. Increased highway safety and better use of travel time are only two of the many advantages of autonomous cars. In order for automated steering to retain the best trajectory despite changes in the road's conditions, lateral steering control is a critical component of autonomous cars. This study aims to investigate the use of a dynamic bicycle model and Stanley controller as a route tracking method, as well as the use of sensors to identify objects and lanes on the way. Discussion will also include the outcomes of MATLAB and SIMULINK tests performed to evaluate these approaches.

Problem solving path planning and path tracking in a 3 DOF hexapod robot using the RRT* algorithm with path optimization and Pose-to-Pose

“…It is still slow and the resulting path is still suboptimal and still not smooth (jagged) [18]. On the other side, Path tracking has several methods, such as Pure Pursuit [19], MPC (Model Predictive Control) [20] and Stanley Controller [21]. However, these path tracking methods are difficult to apply to autonomous robots and need to be adapted to the robot used.…”