Primitive-based Truncated Diffusion for Efficient Trajectory Generation of Differential Drive Mobile Manipulators

2026-04-05

AI Summary

This work addresses the challenge of trajectory planning for differential-drive mobile manipulators in complex 3D environments by proposing a truncated diffusion-based planning framework that integrates keypoint sequence extraction and attention mechanisms. The approach leverages motion primitives to guide the diffusion process, enabling efficient sampling of diverse candidate trajectories from a biased distribution. By incorporating differentiable kinematics and trajectory optimization, the method ensures both dynamic feasibility and task optimality. Experimental results in cluttered 3D simulation scenarios demonstrate that the proposed method significantly outperforms conventional diffusion models and classical planners in terms of planning success rate and trajectory diversity, while maintaining competitive computational efficiency.

Abstract

We present a learning-enhanced motion planner for differential drive mobile manipulators to improve efficiency, success rate, and optimality. For task representation encoder, we propose a keypoint sequence extraction module that maps boundary states to 3D space via differentiable forward kinematics. Point clouds and keypoints are encoded separately and fused with attention, enabling effective integration of environment and boundary states information. We also propose a primitive-based truncated diffusion model that samples from a biased distribution. Compared with vanilla diffusion model, this framework improves the efficiency and diversity of the solution. Denoised paths are refined by trajectory optimization to ensure dynamic feasibility and task-specific optimality. In cluttered 3D simulations, our method achieves higher success rate, improved trajectory diversity, and competitive runtime compared to vanilla diffusion and classical baselines. The source code is released at https://github.com/nmoma/nmoma .

Problem

• trajectory generation
• differential drive mobile manipulators
• motion planning
• diffusion model
• dynamic feasibility

Innovation

• primitive-based diffusion
• truncated diffusion
• keypoint sequence extraction
• attention-based fusion
• trajectory optimization