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

TL;DR

SanD-Planner introduces a sample-efficient diffusion-based local planner that operates in a clamped cubic B-spline control-point space to produce smooth, collision-free trajectories from depth observations. The method decouples trajectory generation from safety verification via an explicit ESDF-based critic, enabling strong performance with only $500$ expert trajectories and zero-shot sim-to-real transfer. Empirical results show state-of-the-art results on InternNav benchmarks, favorable data efficiency, and robust real-world deployment on 2D/3D navigation tasks such as stair traversal. This approach demonstrates that structured trajectory representations and explicit geometric safety checks can dramatically reduce data requirements while maintaining high navigation performance in cluttered environments.

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.

Figures (9)

• Figure 1: Real-world demonstrations of SanD-Planner. Our method enables the robot to traverse cluttered environments with tight clearances, including navigating narrow passages, avoiding obstacles, and climbing stairs.
• Figure 2: Overview of SanD-Planner. The pipeline tokenizes and fuses historical depth images, relative point goals, and previous velocity using a two-layer Transformer encoder. The multi-modal context conditions a diffusion policy to generate B-spline control points through iterative denoising. Finally, an geometric critic module selects the optimal plan from candidates for execution and feeds its initial velocity back to the next planning step to maintain temporal consistency.
• Figure 3: Comparison of trajectory representations among discrete waypoints, cubic spline, and B-spline. (a) Ground-truth A* path (gray) and an 8-waypoint sequence (orange, $0.2m$ spacing). (b) Mean arc-length displacement $\Delta(s)$ under identical perturbations ($10\times$ disturbances). Over the first $1.5m$, the B-spline deviation stays near zero, whereas the cubic spline oscillates. (c, d) We first fit the same ground-truth path with an 8-point interpolating cubic spline (c) and an $8$-point B-spline (d), and then apply the random perturbations to the last four points within a $1m$ radius. Owing to local support and convex hull property, B-splines avoid overshoot and better preserve the global path shape, yielding smaller deviations than cubic splines under the same perturbation range.
• Figure 4: Comparison of depth images. (a) Real-world capture. (b) Clean simulated depth. (c) Domain randomization depth.
• Figure 5: Simulation benchmark environments cai2025navdp. Left: ClutteredEnv provides diverse geometric obstacle layouts. Right: InternScenes features photorealistic indoor settings.