SanD-Planner: Sample-Efficient Diffusion Planner in B-Spline Space for Robust Local Navigation

2026-01-31

AI Summary

This work addresses the limitations of conventional local path planners in highly cluttered and dynamic environments, which often rely heavily on extensive expert demonstrations and exhibit poor generalization. The authors propose a sample-efficient diffusion-based planner that, for the first time, operates in a clamped B-spline space and leverages depth images for imitation learning. By integrating an ESDF-based safety check with explicit collision-avoidance constraints, the method substantially reduces dependence on expert data and alleviates the computational burden of feasibility evaluation. Remarkably, it achieves a 90.1% success rate in simulated cluttered environments and 72.0% in indoor simulations using only 500 training trajectories—merely 0.25% of the baseline dataset size—and demonstrates successful zero-shot transfer to real-world 2D and 3D scenarios.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The challenge of generating reliable local plans has long hindered practical applications in highly cluttered and dynamic environments. Key fundamental bottlenecks include acquiring large-scale expert demonstrations across diverse scenes and improving learning efficiency with limited data. This paper proposes SanD-Planner, a sample-efficient diffusion-based local planner that conducts depth image-based imitation learning within the clamped B-spline space. By operating within this compact space, the proposed algorithm inherently yields smooth outputs with bounded prediction errors over local supports, naturally aligning with receding-horizon execution. Integration of an ESDF-based safety checker with explicit clearance and time-to-completion metrics further reduces the training burden associated with value-function learning for feasibility assessment. Experiments show that training with $500$ episodes (merely $0.25\%$ of the demonstration scale used by the baseline), SanD-Planner achieves state-of-the-art performance on the evaluated open benchmark, attaining success rates of $90.1\%$ in simulated cluttered environments and $72.0\%$ in indoor simulations. The performance is further proven by demonstrating zero-shot transferability to realistic experimentation in both 2D and 3D scenes. The dataset and pre-trained models will also be open-sourced.

Problem

• local navigation
• sample efficiency
• imitation learning
• cluttered environments
• diffusion planner

Innovation

• diffusion planner
• B-spline space
• sample-efficient imitation learning
• ESDF-based safety
• zero-shot transfer