moveit2 integration

The Fanuc CRX-series robots lack an open-source, standards-compliant ROS 2 hardware interface. Method: This paper designs and implements a real-time hardware interface compliant with the ROS 2 Control framework, built upon the R-30iB Plus controller’s communication protocol and integrating an EtherNet/IP/Fieldbus adaptation layer. It supports closed-loop feedback, motion control, trajectory tracking, collision avoidance, and dynamic velocity scaling, and is tightly coupled with MoveIt 2 for high-level motion planning. Contribution/Results: To our knowledge, this is the first modular, open-source ROS 2 hardware interface for the CRX series. It is validated across four representative industrial tasks: command response latency remains stable and bounded; trajectory tracking error stays below joint resolution while respecting velocity constraints; and collision avoidance responses are robust and deterministic. The complete implementation is publicly released under an open-source license, establishing a reusable integration blueprint for industrial robots within the ROS 2 ecosystem.

Existing motion planning frameworks struggle to simultaneously ensure trajectory predictability, cross-platform consistency, and hardware deployability in industrial-grade safety-critical, high-repetition scenarios—such as robotic learning dataset construction and multi-robot coordination. To address this, we propose the first open-source motion planning framework specifically designed for multi-robot manipulation tasks. It integrates search-based algorithmic design, supports major simulators including MuJoCo, SAPIEN, and PyBullet, and provides dual Python/C++ APIs alongside a MoveIt! plugin for seamless hardware integration. Our framework achieves unprecedented reproducibility and cross-platform consistency in trajectory generation for multi-robot tasks. Extensive validation across diverse robot platforms demonstrates significant improvements in planning stability, inter-platform consistency, and deployment efficiency—thereby bridging a critical gap in reliable, safety-aware motion planning tools for real-world industrial applications.

To address the lack of high-performance, scalable real-time drivers for Universal Robots (UR) collaborative manipulators within the ROS 2 ecosystem, this work designs and implements a modular ROS 2 driver built upon the ur_rtde C++ library. The driver establishes a unified real-time data channel enabling millisecond-level state synchronization and command dispatch, natively supports URScript high-level commands, and introduces a novel plugin-based instruction extension mechanism for dynamic injection of custom control logic. A layered architecture decouples communication, control, and interface layers, significantly enhancing configuration flexibility and adaptability across diverse application scenarios. Experimental evaluation demonstrates robust performance in motion planning, multi-node coordination, and complex trajectory execution, along with cross-platform compatibility. The open-source implementation provides a reliable, extensible foundation for UR robot research and deployment in ROS 2 environments.

A significant gap exists between academic research and industrial practice in robot runtime software reconfiguration: while the literature predominantly focuses on structural reconfiguration (e.g., component loading/unloading), industry widely adopts lightweight, low-intrusion parameter-based reconfiguration. Method: We conduct a mixed-methods study comprising a systematic literature review of 78 papers, source-code and documentation analysis of four mainstream robotic frameworks, and empirical evaluation across 48 subsystems plus real-world case studies. Contribution/Results: This work is the first to systematically expose this practice-research divide and establishes the first comprehensive design space for robot reconfiguration. It confirms parameter-level reconfiguration as the sole widely adopted paradigm in practice; identifies critical research gaps and emerging technical trends; and proposes actionable, engineering-oriented improvement pathways. Our findings provide empirically grounded guidance for practitioners in tool selection, framework design, and industry–academia collaboration.

To address the low search efficiency in continuous configuration space for Task and Motion Planning (TAMP), this paper proposes a lightweight neural network framework incorporating kinematic priors, comprising a loosely coupled Chassis Motion Predictor (CMP) and a Full-Body Motion Predictor (FMP). We innovatively embed forward and inverse kinematic constraints into CMP and FMP, respectively, enabling knowledge-guided end-to-end configuration regression; contrastive learning is further introduced to enhance data efficiency. Experiments demonstrate that CMP and FMP achieve configuration prediction accuracies of 96.67% and 98%, respectively, and accelerate motion planning by factors of 24.24× and 153×. Crucially, training data requirements are reduced to only 1/71 and 1/15,052 of those needed by comparable deep learning approaches. The framework thus significantly improves accuracy, computational speed, and generalization capability in TAMP.

This work proposes a multimodal learning framework based on adaptive context fusion to address the limited generalization of existing methods in complex scenarios. The approach dynamically aligns visual and linguistic features and incorporates a lightweight gating mechanism to enable efficient cross-modal integration. Experimental results demonstrate that the model significantly outperforms current state-of-the-art methods across multiple benchmark datasets, achieving improvements of 3.2% in accuracy and 5.7% in robustness. The primary contribution lies in the design of a scalable fusion architecture that effectively mitigates the semantic gap between modalities, offering a novel technical pathway for multimodal understanding tasks.

This study addresses the persistent gap between theoretical control performance and its practical realization in real-world robotic systems, often caused by inadequate discretization, insufficient real-time guarantees, and weak error handling in control software. For the first time from a software engineering perspective, the authors systematically analyze 184 open-source robotic controllers through code review, empirical analysis, and test evaluation, uncovering common deficiencies in application scenarios, implementation details, and verification practices. The findings reveal that most implementations fail to properly account for critical system constraints, and their testing strategies inadequately validate the theoretical assurances they claim. This work highlights a significant disconnect between implementation quality and theoretical promises, offering concrete directions and practical guidelines for developing reliable, verifiable robotic control software.

This work proposes multipanda_ros2, an open-source multi-arm control framework built on ROS 2 to address the challenges of control accuracy, real-time performance, and dynamic consistency in sim-to-real transfer for multi-manipulator systems. The framework enables single-process real-time control of an arbitrary number of Franka arms, integrates high-fidelity MuJoCo simulation, and enhances force/torque accuracy through inertial parameter identification. It introduces a novel controllet-feature design pattern that achieves controller switching latency of ≤2 ms and extends soft robotics methodologies to rigid dual-arm contact tasks, establishing quantitative metrics to bridge the sim-to-real gap. Experimental results demonstrate stable 1 kHz torque control and significantly improved consistency between simulation and physical execution, offering a reproducible, high-fidelity platform for complex multi-arm cooperative tasks.

This work addresses the limited generalizability and transferability of current foundation models in robotics, which often require custom integration of perception, actuation, and safety mechanisms. To overcome this, the authors propose ROSClaw—a model-agnostic execution layer that enables plug-and-play deployment of arbitrary foundation models on any ROS 2 robot by integrating the OpenClaw agent runtime with ROS 2. Key innovations include standardized capability discovery, multimodal observation normalization, action validation within configurable safety bounds, and structured audit logging. Experiments across three robotic platforms and four foundation models demonstrate up to a 4.8× difference in non-policy action proposal rates and show that the proposed execution layer significantly improves task success rates and safety across diverse frameworks.