Geometric path tracking algorithm for autonomous driving in pedestrian environment
Authors:
- 1 July 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1669-1674
Abstract
This paper proposes an alternative formulation to the pure pursuit path tracking algorithm for autonomous driving. The current approach has tendencies to cut corners, and therefore results in poor path tracking accuracy. The proposed method considers not only the relative position of the pursued point, but also the orientation of the path at that point. A steering control law is designed in accordance with the kinematic equations of motion of the vehicle. The effectiveness of the algorithm is then tested by implementing it on an autonomous golf cart, driving in a pedestrian environment. The experimental result shows that the new algorithm reduces the root mean square (RMS) cross track error for the same given pre-programmed path by up to 46 percent, while having virtually no extra computational cost, and still maintaining the chatter free property of the original pure pursuit controller.Keywords
This publication has 7 references indexed in Scilit:
- Autonomous golf cars for public trial of mobility-on-demand service2015
- Development of lateral control system for autonomous vehicle based on adaptive pure pursuit algorithm2014
- Mapping with synthetic 2D LIDAR in 3D urban environment2013
- Autonomous driving in urban environments: Boss and the Urban ChallengeJournal of Field Robotics, 2008
- Motion Planning in Complex Environments Using Closed-loop Prediction2008
- Autonomous Automobile Trajectory Tracking for Off-Road Driving: Controller Design, Experimental Validation and RacingAmerican Control Conference (ACC), 2007
- Fuzzy Control for Autonomous Ground Vehicles2000