Bench-Push: Benchmarking Pushing-based Navigation and Manipulation Tasks for Mobile Robots

2025-12-12

AI Summary

Mobile robots often fail to complete tasks in cluttered environments containing movable objects due to prohibitions on physical interaction, while existing push-based navigation research lacks a unified, reproducible evaluation benchmark. To address this, we propose the first simulation benchmark for joint navigation and manipulation via pushing, featuring diverse scenarios—including maze traversal, icy-surface vessel navigation, box delivery, and area clearing. Our contribution includes a novel multi-dimensional evaluation framework supporting partial-success assessment, quantification of interaction energy consumption, and joint measurement of task efficiency. We provide a modular, open-source Python library built on PyBullet and Gazebo, integrating dynamics-aware reward functions and a cross-environment baseline evaluation toolkit. Experiments demonstrate significant performance disparities among state-of-the-art algorithms on pushing tasks, markedly improving methodological comparability. The codebase, comprehensive documentation, and pretrained models are publicly released and widely adopted by the community.

Technology Category

• Intelligent Robots: Manipulation
• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Mobile robots are increasingly deployed in cluttered environments with movable objects, posing challenges for traditional methods that prohibit interaction. In such settings, the mobile robot must go beyond traditional obstacle avoidance, leveraging pushing or nudging strategies to accomplish its goals. While research in pushing-based robotics is growing, evaluations rely on ad hoc setups, limiting reproducibility and cross-comparison. To address this, we present Bench-Push, the first unified benchmark for pushing-based mobile robot navigation and manipulation tasks. Bench-Push includes multiple components: 1) a comprehensive range of simulated environments that capture the fundamental challenges in pushing-based tasks, including navigating a maze with movable obstacles, autonomous ship navigation in ice-covered waters, box delivery, and area clearing, each with varying levels of complexity; 2) novel evaluation metrics to capture efficiency, interaction effort, and partial task completion; and 3) demonstrations using Bench-Push to evaluate example implementations of established baselines across environments. Bench-Push is open-sourced as a Python library with a modular design. The code, documentation, and trained models can be found at https://github.com/IvanIZ/BenchNPIN.

Problem

• Benchmarking pushing-based navigation and manipulation for mobile robots
• Addressing reproducibility and cross-comparison limitations in existing evaluations
• Providing simulated environments, metrics, and baselines for pushing tasks

Innovation

• Unified benchmark for pushing-based robot navigation
• Simulated environments with varying complexity levels
• Novel evaluation metrics for interaction efficiency