Learning to Tune Pure Pursuit in Autonomous Racing: Joint Lookahead and Steering-Gain Control with PPO
2026-02-20 • Robotics
RoboticsArtificial IntelligenceMachine Learning
AI summaryⓘ
The authors improved a common method called Pure Pursuit (PP), used in self-driving race cars to follow paths smoothly. Usually, key settings in PP like how far ahead the car looks and how much it turns are fixed or roughly adjusted, which doesn't work well across different tracks. They used a type of machine learning called reinforcement learning to adjust these settings on the fly based on the car's speed and road curves. Tested in simulations and a real car, their approach worked better than traditional PP and other advanced controls, making driving smoother and more accurate without needing to retune for each track.
Pure PursuitReinforcement LearningProximal Policy OptimizationLookahead DistanceSteering GainPath TrackingAutonomous RacingROS 2Kinematic MPCF1TENTH Gym
Authors
Mohamed Elgouhary, Amr S. El-Wakeel
Abstract
Pure Pursuit (PP) is widely used in autonomous racing for real-time path tracking due to its efficiency and geometric clarity, yet performance is highly sensitive to how key parameters-lookahead distance and steering gain-are chosen. Standard velocity-based schedules adjust these only approximately and often fail to transfer across tracks and speed profiles. We propose a reinforcement-learning (RL) approach that jointly chooses the lookahead Ld and a steering gain g online using Proximal Policy Optimization (PPO). The policy observes compact state features (speed and curvature taps) and outputs (Ld, g) at each control step. Trained in F1TENTH Gym and deployed in a ROS 2 stack, the policy drives PP directly (with light smoothing) and requires no per-map retuning. Across simulation and real-car tests, the proposed RL-PP controller that jointly selects (Ld, g) consistently outperforms fixed-lookahead PP, velocity-scheduled adaptive PP, and an RL lookahead-only variant, and it also exceeds a kinematic MPC raceline tracker under our evaluated settings in lap time, path-tracking accuracy, and steering smoothness, demonstrating that policy-guided parameter tuning can reliably improve classical geometry-based control.