In this paper, a lateral controller is proposed for an over-actuated vehicle. The controller is formulated as a linear time-varying model predictive controller. The aim of the controller is to track a desired path smoothly, by making use of the vehicle crabbing capability (sideways movement) and minimizing the magnitude of curvature used. To do this, not only the error to the path is minimized, but also the error to the desired orientation and the control signals requests. The controller uses an extended kinematic model that takes into consideration the vehicle crabbing capability and is able to track not only kinematically feasible paths, but also plan and track over non-feasible discontinuous paths. Ackermann steering geometry is used to transform the control requests, curvature, and crabbing angle, to wheel angles. Finally, the controller performance is evaluated first by simulation and, after, by means of experimental tests on an over-actuated autonomous research vehicle.
Lateral Model Predictive Control for Over-Actuated Autonomous Vehicle
2017-06-01
2901741 byte
Conference paper
Electronic Resource
English
Lateral Model Predictive Control for Over-Actuated Autonomous Vehicle
British Library Conference Proceedings | 2017
|Model Predictive Control for Autonomous Driving of Over-Actuated Research Vehicle
BASE | 2016
|