Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Integrating Active Sensing and Rearrangement Planning for Efficient Object Retrieval from Unknown, Confined, Cluttered Environments

Junyong Kim, Hanwen Ren, Ahmed H. Qureshi

TL;DR

A new, integrated heuristic-based active sensing and Monte-Carlo Tree Search (MCTS)-based retrieval planning approach that outperforms baseline methods by a significant margin in terms of the success rate, the object rearrangement planning time consumption and the number of planning trials before successfully retrieving the target.

Abstract

Retrieving target objects from unknown, confined spaces remains a challenging task that requires integrated, task-driven active sensing and rearrangement planning. Previous approaches have independently addressed active sensing and rearrangement planning, limiting their practicality in real-world scenarios. This paper presents a new, integrated heuristic-based active sensing and Monte-Carlo Tree Search (MCTS)-based retrieval planning approach. These components provide feedback to one another to actively sense critical, unobserved areas suitable for the retrieval planner to plan a sequence for relocating path-blocking obstacles and a collision-free trajectory for retrieving the target object. We demonstrate the effectiveness of our approach using a robot arm equipped with an in-hand camera in both simulated and real-world confined, cluttered scenarios. Our framework is compared against various state-of-the-art methods. The results indicate that our proposed approach outperforms baseline methods by a significant margin in terms of the success rate, the object rearrangement planning time consumption and the number of planning trials before successfully retrieving the target. Videos can be found at https://youtu.be/tea7I-3RtV0.

Integrating Active Sensing and Rearrangement Planning for Efficient Object Retrieval from Unknown, Confined, Cluttered Environments

TL;DR

A new, integrated heuristic-based active sensing and Monte-Carlo Tree Search (MCTS)-based retrieval planning approach that outperforms baseline methods by a significant margin in terms of the success rate, the object rearrangement planning time consumption and the number of planning trials before successfully retrieving the target.

Abstract

Retrieving target objects from unknown, confined spaces remains a challenging task that requires integrated, task-driven active sensing and rearrangement planning. Previous approaches have independently addressed active sensing and rearrangement planning, limiting their practicality in real-world scenarios. This paper presents a new, integrated heuristic-based active sensing and Monte-Carlo Tree Search (MCTS)-based retrieval planning approach. These components provide feedback to one another to actively sense critical, unobserved areas suitable for the retrieval planner to plan a sequence for relocating path-blocking obstacles and a collision-free trajectory for retrieving the target object. We demonstrate the effectiveness of our approach using a robot arm equipped with an in-hand camera in both simulated and real-world confined, cluttered scenarios. Our framework is compared against various state-of-the-art methods. The results indicate that our proposed approach outperforms baseline methods by a significant margin in terms of the success rate, the object rearrangement planning time consumption and the number of planning trials before successfully retrieving the target. Videos can be found at https://youtu.be/tea7I-3RtV0.

Paper Structure

This paper contains 15 sections, 5 equations, 2 figures, 2 tables, 1 algorithm.

Table of Contents

  1. INTRODUCTION
  2. Related Work
  3. Methods
  4. Multi-objective Active Sensing
  5. Scene Coverage Sensing
  6. Region-specific Sensing
  7. Integrated Sensing and Object Retrieval Planning
  8. Retrieval Reachability Sensing and Planning
  9. Retrieval Rearrangement Sensing and Planning
  10. Execution Algorithm
  11. Experiments and Discussions
  12. Effectiveness of integrated active sensing and planning
  13. Evaluation of Object Retrieval Planning:
  14. Real Robot Experiments
  15. Conclusions and Future Work

Figures (2)

  • Figure 1: Depiction of cluttered example testing environments.
  • Figure 2: Our method, MAS+OR-MCTS, solves the object retrieval task in the confined, real-world cabinet environment shown in (a) where the target object banana is surrounded by 5 sizeable cylindrical movable obstacles. The approach starts by actively sensing the environment to (b) detect the target object, (c) observe the robot swept volumes for it to be able to begin running OR-MCTS for target retrieval, and (d) other critical areas based on the FAS + OR-MCTS. Once a feasible object rearrangement plan is found, the robot executes it (e-g), resulting in successful retrieval of the target object (h).