SPADE: Sketch-guided Path Planning Augmented with Diffusion Experts
2026-06-02 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors improved how autonomous robots plan paths by teaching them from example behaviors. They created a new tool and used a special training method involving diffusion-based augmentation to make the robots better at handling new places and mistakes in learning. Their method works more accurately and efficiently than previous ones, needing fewer parameters while still running fast enough for real-time use. They tested their approach with expert data to show it can generalize well without heavy hardware needs.
Autonomous Mobile RobotsPath PlanningImitation LearningBehavioral CloningDiffusion-Based AugmentationROS 2Absolute Pose ErrorFréchet Inception DistanceGeneralizationReal-time Systems
Authors
Charbel Abi Hana, Tatiana Ghantous, Mikael Khalil, Anthony Rizk
Abstract
Path planning is essential for Autonomous Mobile Robots (AMRs). Conventional methods for incorporating human preferences into planning typically rely on either complex reward engineering or hardware-intensive solutions. Recent state-of-the-art frameworks leverage imitation learning to train behavior-specific path planning models from expert demonstrations. However, these approaches face two key limitations: limited generalization to unseen environments and low robustness in demonstration collection. To address these challenges, this work introduces an enhanced framework that focuses on two main contributions: an overhauled annotation tool built on ROS 2, and a novel training strategy that integrates diffusion-based augmentation into baseline behavioral cloning models. A dataset of expert demonstrations is provided and evaluated through ablation studies to assess the robustness of the proposed solution. The enhanced approach outperforms state-of-the-art methods with 39.1% lower Absolute Pose Error (APE) and 33.5% lower Fr'echet Inception Distance (FID) while having 93.8% less trainable parameters. Moreover it attains diffusion-level generalization while preserving the real-time, on-edge properties of state-of-the-art models.