Jul 2026· 2026 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)· pp. 1-6· 0 citations· 18 references
Abstract
Mobile manipulators on construction sites offer considerable potential for increasing productivity, as the transport of materials and the execution of precise assembly work can be increasingly automated. However, the coordination of several such robots is a complex planning task, as task assignment, navigation, and reachability planning must be solved simultaneously and under dynamic environmental conditions. This work presents a multi-agent reinforcement learning (RL) approach that enables multiple mobile manipulators to complete a set of tasks in a structured environment. Each agent makes decentralized decisions about task selection and navigation, with kinematic reachability ensured by an integrated inverse kinematic solver. The policy is trained using proximal policy optimization (PPO), supported by a reward function that encourages both navigation progress and efficient task distribution. Simulation results show that the trained model is able to efficiently distribute tasks among multiple robots while taking kinematic constraints into account. The proposed method is superior to a greedy baseline that selects the nearest available task. With four robots and 35 tasks, the multi-agent RL approach achieves a success rate of 100%, while the baseline reaches only 55%.
Reinforcement learning (RL) has shown considerable promise for robotic decision-making, yet deploying multi-agent RL (MARL) on physical multi-robot systems in industrial environments remains challenging. This paper investigates the real-world applicability of decentralized MARL for multi-robot multi-machine tending. We propose Feature-fusion Multi-Agent Proximal Policy Optimization (FMAPPO), which fuses 2D LiDAR measurements with task-specific state information to enable safe decentralized multi-robot task assignment and navigation. A complete simulation-to-reality pipeline was developed using high-fidelity robotic simulation and ROS2 and deployed on physical mobile-manipulator platforms operating under realistic real-world conditions, with the robotic arms disabled during the experiments. We further investigate the sensitivity of the learned policy to command update frequency, an important consideration for real-world deployment. Comparative evaluation in simulation demonstrated that FMAPPO significantly outperformed state-of-the-art baselines with a large effect size, achieving improvements of 106\% and 21\% in parts delivery and 48\% and 11\% in parts collection over MAPPO and SMAPPO, respectively. FMAPPO also increased machine utilization by 31 and 10 percentage points, respectively, while reducing collisions by 18\% and 15\% and increasing the safety score by 14 and 6 percentage points compared with MAPPO and SMAPPO, respectively. Furthermore, real-world experiments demonstrated that the learned decentralized policies can coordinate multiple robots to service multiple machines while maintaining safe operation under real-world sensing and control constraints. Videos of the real-world experiment are available online https://anonymouspapers123.github.io/FMAPPO/.
A. Abdalwhab, Giovanni Beltrame, David St-Onge· 0 citations
Robotic manipulators operating in cluttered environments require collision-free trajectories that remain executable under kinematic and dynamic constraints. This paper proposes a reinforcement learning (RL)-guided multi-objective trajectory planning framework, termed RL-MOP-HNE, for a 6-DOF UR5 manipulator. The planning model simultaneously minimizes path length, energy consumption, and execution time while satisfying collision-avoidance, kinematic, and dynamic constraints. A tabular SARSA agent is embedded into the evolutionary search to adaptively select search behaviours according to the current optimization state. To improve the balance between exploration and exploitation, a Gaussian-perturbation adaptive hybrid crossover operator is integrated with a hierarchical neighborhood evolution (HNE) strategy, enabling progressive population refinement throughout the search process. The proposed method is evaluated in three representative environments with increasing planning complexity, including single-obstacle, narrow three-obstacle, and irregular five-obstacle scenarios, and is compared with MOEA/D, MOPSO, MSCLPSO, NSGA-II, and RL-NSGA-II. Experimental results show that RL-MOP-HNE generates feasible trajectories in all test cases and achieves the lowest dynamic-stability-prioritized composite scores among the compared algorithms. The planned trajectories exhibit smoother joint motion and lower velocity fluctuations, although these improvements are generally accompanied by longer execution times. Complementary analyses, including time scaling, manipulability, clearance evaluation, statistical significance tests, and ablation studies, further explain the performance characteristics of the proposed framework and quantify the contribution of its key components. The proposed framework is therefore well suited to robotic applications where motion stability and dynamic executability are of greater importance than minimum-time operation.
Zhen-Long Zhao, Shu-Tao Hao, Bi-Hao Jin et al.· Scientific Reports· 0 citations
A softmin-based adaptive blending mechanism that automatically selects the most suitable path from the champion solutions on the Pareto front according to regional environmental conditions, thereby eliminating the need for manual user intervention is introduced.
Osman Emre Turan, Oğuz Mısır, Mustafa Özden· Measurement science and tech...· 0 citations
: This paper provides a thorough survey and integrative presentation of cooperative path planning for multi-robot systems operating in dynamic, cluttered, and partially observable environments. People synthesise algorithmic foundations ranging from heuristic graph search to sampling-based motion planners, including A*, D* Lite, and Safe Interval Path Planning for discrete/time-augmented spaces, as well as RRT, RRT*, and Informed RRT* for continuous configuration spaces. Multi-agent coordination techniques are reviewed, covering reciprocal collision avoidance (ORCA) and centralised Multi-Agent Path Finding (MAPF) solvers such as Conflict-Based Search (CBS) and bounded-suboptimal variants (ECBS). The paper also examine control and safety layers like Model Predictive Control and Control Barrier Functions that translate plans into dynamically feasible commands with safety guarantees. Recent progress in cooperative multi-agent reinforcement learning (MAPPO, QMIX, MADDPG) is evaluated for adaptability under partial observability and nonstationary environments. Applications in warehousing, intelligent transportation, and disaster response are used to illustrate practical trade-offs and integration patterns, referencing real-world systems such as Kiva-style warehouse fleets and autonomous driving pipelines. The paper concludes with a focused discussion on open challenges — scalability with guarantees, safety under uncertainty, sim-to-real transfer, and planning – control interface fragility — and proposes research directions including learning-augmented heuristics, unified safety-aware planning, adaptive MPC – CBF filters, and more informative benchmarks to drive reproducible progress.
Yun Pan· Proceedings of the 3rd Inter...· 0 citations
Modern manufacturing faces increasing demands for flexibility, customization, and productivity under dynamic conditions. Multi-robot systems offer a promising solution by enabling cooperative execution of complex tasks, such as assembly and cooperative manipulation. In this context, Multi-Agent Reinforcement Learning (MARL) has emerged as a promising paradigm to enhance coordination and adaptability in industrial settings. MARL enables multiple agents to learn and interact in shared environments to achieve common goals within complex and dynamic industrial processes. In this paper, a deep analysis of MARL applied to industrial multi-robot systems based on a systematic review is presented, with particular focus on cooperative manipulation tasks. Following PRISMA guidelines, we analyze a total of 30 articles published between 2016 and 2026, selected independently by two of the authors from an initial pool of 102 records retrieved from Scopus and Web of Science. These articles were used to address five key questions regarding MARL algorithms, control architectures, industrial applications and validation practices. These research questions seek to examine gaps and trends at the research level which are important for the development of multi-agent control technologies. This review shows a clear prevalence of model-free algorithms under Centralized Training with Decentralized Execution (CTDE) architectures, with validation mainly performed in simulation. Despite promising results and high potential for impact, critical gaps remain in scalability, reproducibility, and sim-to-real transfer, limiting real deployment in manufacturing environments. To address these challenges and fill current gaps, we outline actionable research directions, such as hybrid MARL approaches, standardized industrial benchmarks, digital twin pipelines, and safety-aware deployment strategies, to accelerate MARL adoption in industrial environments.
Francisco J. Huertos, Oihane Bañales, Pedro Álvarez et al.· Robotics· 0 citations
Mobile robots often face challenges operating in partially known environments where success depends on factors such as robot capabilities, task characteristics, and the environment. This letter presents a generalized, modular framework that enables robots to explicitly account for these factors through the concept of a mission-gain: a pose dependent metric quantifying mission performance (e.g. visibility). When the environment can be proactively explored, a cost map encoding the value of the mission-gain is generated and integrated into a cost function for a model predictive path integral (MPPI) controller, which supports arbitrary mission-specific cost functions. The controller generates inputs that balance progress toward the goal with improved mission performance. The map generation is decoupled from the control execution, making the framework modular and well-suited for heterogeneous multi-robot systems. To address transmission demands of large maps, we distill the map into a Gaussian mixture model (GMM) approximation, yielding a representation with far fewer parameters. The approach is validated in simulations and lab experiments across two scenarios: occlusion-aware navigation and vibration minimization over rough-terrain. The results for both scenarios show the mission-gain aware robot successfully reaching the goal while improving mission safety.
Patrick Sherman, N. Bezzo· IEEE Robotics and Automation...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
