Search papers, labs, and topics across Lattice.
This paper introduces DiffuSearch, a hybrid trajectory planner for autonomous driving that integrates a unified set of objectives across both trajectory generation and refinement stages. By employing a guided diffusion model for initial trajectory prediction and a Monte Carlo Tree Search (MCTS) for local refinement, the approach ensures consistency in driving goals such as collision avoidance and comfort. Experimental results on benchmark datasets reveal that DiffuSearch significantly reduces collisions and enhances comfort, particularly in complex scenarios, with MCTS refinement identified as a key contributor to its performance gains.
Achieving state-of-the-art performance in trajectory planning, DiffuSearch reduces collisions and enhances comfort by aligning objectives across generation and refinement stages.
In trajectory planning for autonomous driving, hybrid planning architectures are often realized as a collection of disparate modules, each with its own objectives. This lack of a unifying principle can lead to inconsistencies between the initial and refined trajectory, resulting in suboptimal behavior. We address this by introducing DiffuSearch, a novel hybrid planner that uses a unified set of objectives across generation and refinement. Our model encourages all components to follow the same shared driving goals: collision avoidance, drivable area compliance, comfort, and progress. DiffuSearch employs a two-stage architecture. First, a guided diffusion model generates a scene-consistent, joint trajectory prediction, using our driving objectives as differentiable guidance functions to implicitly steer the denoising process. Second, a Monte Carlo Tree Search (MCTS) in a discretized action space performs an explicit, local refinement of this proposal, leveraging the same driving objectives as its reward function. This synergistic design leverages the diffusion model's strength in finding scene-consistent solutions combined with the explainable, constraint-aware refinement of MCTS. Experiments on nuPlan and interPlan reactive closed-loop benchmarks demonstrate that DiffuSearch achieves strong and often state-of-the-art performance, substantially reducing collisions and improving comfort, particularly in complex, interactive scenarios. Our ablation studies indicate that MCTS refinement is the main mechanism behind the gains, while sharing objectives between implicit guidance and explicit search provides further consistent improvements.
