Curriculum reinforcement learning-based drifting along a general path for autonomous vehicles

Expert drivers possess the ability to execute high sideslip angle maneuvers, commonly known as drifting, duringracing to navigate sharp corners and execute rapid turns. However, existing model-based controllers encounter chal-lenges in handling the highly nonlinear dynamics associated with drifting along general paths. While reinforcementlearning-based methods alleviate the reliance on explicit vehicle models, training a policy directly for autonomousdrifting remains difficult due to multiple objectives. In this paper, we propose a control framework for autonomousdrifting in the general case, based on curriculum reinforcement learning. The framework empowers the vehicleto follow paths with varying curvature at high speeds, while executing drifting maneuvers during sharp corners.Specifically, we consider the vehicle's dynamics to decompose the overall task and employ curriculum learning tobreak down the training process into three stages of increasing complexity. Additionally, to enhance the generaliza-tion ability of the learned policies, we introduce randomization into sensor observation noise, actuator action noise,and physical parameters. The proposed framework is validated using the CARLA simulator, encompassing variousvehicle types and parameters. Experimental results demonstrate the effectiveness and efficiency of our framework inachieving autonomous drifting along general paths. The code is available athttps://github.com/BIT-KaiYu/drifting