Closed-loop control improves learned driving planner safety and completion rates
Closed-Loop Refinement and Execution for Learned Driving Planners
Robotics
Summary
Driving planners that learn from data often work well when tested on old routes, but can still cause the car to stop unexpectedly or get too close to other cars when actually driving. The authors present CLRE, a two-layer system that takes a planned path and adjusts it continually to avoid crashes and unsafe moves without changing the original planner. In simulations, CLRE made the car drive further along routes, reduce collisions, and achieve higher overall safety scores.
What this means in practice
- •For autonomous vehicle engineers: Enhance the safety and reliability of learned driving planners by adding a real-time control layer that refines trajectories and ensures collision-free execution.
- •For robotics control teams: Use hierarchical receding-horizon control to improve execution of planned movements in dynamic environments while ensuring safety constraints.
Authors
Huaijin Hu, Shanting Wang, Zhongyu Mo, Andreas A. Malikopoulos
Abstract
Learning-based driving planners are usually trained and evaluated in open loop against logged trajectories. In closed loop, a trajectory with small displacement error can still stall the vehicle, steer it into a conflict with surrounding agents, or be executed with abrupt braking. We introduce Closed-Loop Refinement and Execution (CLRE), a hierarchical receding-horizon control framework designed to mitigate these failure modes while leaving the upstream planner frozen and adding no new learned model. The upper layer treats the nominal trajectory as a reference and solves a finite-horizon optimal control problem that trades route progress against interaction with predicted agents. Solving it from several initializations gives a candidate set, and a prediction-conditioned oriented-bounding-box (OBB) feasibility test retains only candidates whose minimum predicted OBB clearance over the horizon meets a threshold. The lower layer executes the lowest-cost survivor, or a route-centerline backup when none remains, through the tracking controller supplied with the planner, augmented by a range-based speed bound and a saturated proportional braking law. In closed-loop simulation on 126 Bench2Drive routes with VAD as the upstream planner, CLRE raises the driving score from 43.41 to 56.42 and route completion from 57.27 to 72.23, and reduces collision events from 70 to 53.