Interpretable Goal-based Prediction and Planning for Autonomous Driving

We propose an integrated prediction and planning system for autonomous driving which uses rational inverse planning to recognise the goals of other vehicles. Goal recognition informs a Monte Carlo Tree Search (MCTS) algorithm to plan optimal maneuvers for the ego vehicle. Inverse planning and MCTS utilise a shared set of defined maneuvers and macro actions to construct plans which are explainable by means of rationality principles. Evaluation in simulations of urban driving scenarios demonstrate the system's ability to robustly recognise the goals of other vehicles, enabling our vehicle to exploit non-trivial opportunities to significantly reduce driving times. In each scenario, we extract intuitive explanations for the predictions which justify the system's decisions.

Automated summary

The proposed integrated prediction and planning system for autonomous driving utilizes rational inverse planning to recognize other vehicles' goals, enabling an optimal maneuver planning for the ego vehicle through a Monte Carlo Tree Search (MCTS) algorithm. The system demonstrates the ability to reduce driving times significantly in urban driving scenarios by robustly recognizing other vehicles' goals, with explanations for the predictions based on rationality principles.