Learning Model Predictive Control for Competitive Autonomous Racing
This work addresses real-time competitive racing for autonomous agents, representing an incremental improvement over existing single-agent formulations.
The paper tackles the problem of enabling competitive multi-agent autonomous racing by addressing limitations in single-agent learning model predictive control (LMPC), such as lack of state space exploration and non-convex terminal sets, resulting in a method that explores state space through multiple initializations and maintains convexity in the terminal safe set.
The goal of this thesis is to design a learning model predictive controller (LMPC) that allows multiple agents to race competitively on a predefined race track in real-time. This thesis addresses two major shortcomings in the already existing single-agent formulation. Previously, the agent determines a locally optimal trajectory but does not explore the state space, which may be necessary for overtaking maneuvers. Additionally, obstacle avoidance for LMPC has been achieved in the past by using a non-convex terminal set, which increases the complexity for determining a solution to the optimization problem. The proposed algorithm for multi-agent racing explores the state space by executing the LMPC for multiple different initializations, which yields a richer terminal safe set. Furthermore, a new method for selecting states in the terminal set is developed, which keeps the convexity for the terminal safe set and allows for taking suboptimal states.