PaiP: An Operational Aware Interactive Planner for Unknown Cabinet Environments
chengjin wang, zheng yan, yanmin Zhou, Runjie Shen, Zhipeng Wang, Bin Cheng, Bin He
AI summary
Problem
Traditional collision-free planners fail in visually occluded, narrow cabinet environments with stacked objects, often causing catastrophic collisions or getting stuck in unsolvable free spaces. Robots lack the ability to autonomously perceive and leverage physical interaction features to navigate or reconfigure these constrained spaces in real time.
Approach
PaiP uses a real-time closed-loop framework that fuses tactile and proprioceptive data to infer object mechanical properties (modeled as spring-damper-friction systems) and converts them into an operational cost map, which guides an extended sampling-based planner to optimize both path and interaction costs.
Key results
- Closed-loop tactile perception-action framework for real-time interactive planning
- Constitutive spring-damper-friction model for autonomous extraction of object interaction features
- Continuous operational cost map integrating mechanical properties into sampling-based planners
- Demonstrated robust real-time navigation and obstacle reconfiguration in narrow, partially observable cabinet environments
Why it matters
Enables robots to safely and adaptively manipulate objects in cluttered, visually occluded domestic and industrial storage spaces where traditional planners fail.
Abstract
Box/cabinet scenarios with stacked objects pose significant challenges for robotic motion due to visual occlusions and constrained free space. Traditional collision-free trajectory planning methods often fail when no collision-free paths ex- ist, and may even lead to catastrophic collisions caused by invisible objects. To overcome these challenges, we propose an operational aware interactive motion planner (PaiP) a real-time closed-loop planning framework utilizing multimodal tactile perception. This framework autonomously infers object interaction features by perceiving motion effects at interaction interfaces. These interaction features are incorporated into grid maps to generate operational cost maps. Building upon this representation, we extend sampling-based planning methods to interactive planning by optimizing both path cost and operational cost. Experimental results demonstrate that PaiP achieves robust motion in narrow spaces. Project page: https: //travelers-lab.github.io/PaiP/