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

TL;DR

This paper introduces a learning-enhanced motion planner using primitive-based truncated diffusion for efficient and diverse trajectory generation of differential drive mobile manipulators, with improved success rates and runtime.

Contribution

It proposes a novel primitive-based truncated diffusion model combined with keypoint encoding and attention fusion for better trajectory sampling and planning.

Findings

Higher success rate in cluttered 3D environments

Improved trajectory diversity over baseline methods

Competitive runtime performance

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 .