RaceMOP: Mapless Online Path Planning for Multi-Agent Autonomous Racing using Residual Policy Learning
The author introduces RaceMOP, a mapless online path planning method for multi-agent racing, combining an artificial potential field planner with residual policy learning to improve decision-making capabilities in autonomous racing.
Learning Model Predictive Control with Error Dynamics Regression for Autonomous Racing
The author proposes a novel LMPC strategy for autonomous racing that focuses on learning error dynamics to improve robustness and performance. By combining a nominal model with local linear data-driven learning, the approach aims to explore handling limits incrementally and safely.
Unifying F1TENTH Autonomous Racing: Survey, Methods, and Benchmarks
The author aims to unify the field of F1TENTH autonomous racing by surveying current approaches, describing common methods, and providing benchmark results to facilitate clear comparison and establish a baseline for future work.
© 2024 by Linnk AI