• Corpus ID: 234763080

Graph Neural Networks for Decentralized Multi-Robot Submodular Action Selection

  title={Graph Neural Networks for Decentralized Multi-Robot Submodular Action Selection},
  author={Lifeng Zhou and Vishnu Dutt Sharma and Qingbiao Li and Amanda Prorok and Alejandro Ribeiro and Vijay R. Kumar},
In this paper, we develop a learning-based approach for decentralized submodular maximization. We focus on applications where robots are required to jointly select actions, e.g., motion primitives, to maximize team submodular objectives with local communications only. Such applications are essential for large-scale multi-robot coordination such as multi-robot motion planning for area coverage, environment exploration, and target tracking. But the current decentralized submodular maximization… 

Figures and Tables from this paper

Learning Control Admissibility Models with Graph Neural Networks for Multi-Agent Navigation

A new learning-based approach to multi-agent navigation that shifts the focus from learning the optimal control policy to a set-theoretic representation of admissible control policies to achieve compositional inference.

A Framework for Real-World Multi-Robot Systems Running Decentralized GNN-Based Policies

This work creates a framework based on ROS2 and elaborate its details in this paper, and presents first-of-a-kind results that show successful real-world deployment of GNN-based policies on a decentralized multi-robot system relying on Adhoc communication.

D2CoPlan: A Differentiable Decentralized Planner for Multi-Robot Coverage

D2C O P LAN can be seamlessly combined with other learning methods to learn end-to-end, resulting in a better solution than the individually trained modules, opening doors to further research for tasks that remain elusive with classical methods.

Graph neural networks for decentralized multi-agent perimeter defense

This work uses graph neural networks (GNNs) to develop an imitation learning framework that learns a mapping from defenders’ local perceptions and their communication graph to their actions that performs closer to the expert algorithm and is superior to other baseline algorithms by capturing more intruders.

Scalable Multi-Agent Reinforcement Learning through Intelligent Information Aggregation

InforMARL is proposed, a novel architecture for multi-agent reinforcement learning (MARL) which uses local information intelligently to compute paths for all the agents in a decentralized manner and scales well to environments with arbitrary numbers of agents and obstacles.

Learning Decentralized Strategies for a Perimeter Defense Game with Graph Neural Networks

This work designs a graph neural network-based learning frame- work to learn a mapping from defenders' local perceptions and the communication graph to defenders’ actions such that the learned actions are close to that generated by a centralized expert algorithm.

Learning-based Motion Planning in Dynamic Environments Using GNNs and Temporal Encoding

A GNN-based approach that uses temporal encoding and imitation learning with data aggregation for learning both the em-beddings and the edge prioritization policies that can accelerate online planning over state-of-the-art complete dynamic planning algorithms.

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

This work describes a loop closure module that is able to prioritize which loop closures to compute based on the underlying pose graph, the proximity to known beacons, and the characteristics of the point clouds.

SubMo-GNN: Property- and Structure-Aware Diverse Molecular Selection with Representation Learning and Mathematical Diverse-Selection Framework

The proposed method enables researchers to obtain diverse sets of molecules for discovering new molecules and novel chemical reactions, and the proposed diversity criterion is useful for discussing the diversity of molecular libraries from a new property-based perspective.

Selecting molecules with diverse structures and properties by maximizing submodular functions of descriptors learned with graph neural networks

The proposed method enables researchers to obtain diverse sets of molecules for discovering new molecules and novel chemical reactions, and the proposed diversity criterion is useful for discussing the diversity of molecular libraries from a new property-based perspective.



Graph Neural Networks for Decentralized Multi-Robot Path Planning

A combined model is proposed that automatically synthesizes local communication and decision-making policies for robots navigating in constrained workspaces that shows its capability to generalize to previously unseen cases.

Learning Scheduling Policies for Multi-Robot Coordination With Graph Attention Networks

This work develops a novel graph attention network-based scheduler to automatically learn features of scheduling problems towards generating high-quality solutions, which significantly outperforms prior state-of-the-art, approximate methods.

Message-Aware Graph Attention Networks for Large-Scale Multi-Robot Path Planning

The Message-Aware Graph Attention neTwork (MAGAT) is based on a key-query-like mechanism that determines the relative importance of features in the messages received from various neighboring robots and is able to achieve a performance close to that of a coupled centralized expert algorithm.

Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents

This work appears to be the first study of fully decentralized MARL algorithms for networked agents with function approximation, with provable convergence guarantees, and can be implemented in an online fashion.

Distributed Attack-Robust Submodular Maximization for Multi-Robot Planning

This paper proposes a general-purpose distributed algorithm towards robust optimization at scale, with local communications only, and name it distributed robust maximization (DRM), which proposes a divide-and-conquer approach that distributively partitions the problem among K cliques of robots.

Distributed matroid-constrained submodular maximization for multi-robot exploration: theory and practice

Real-time simulation and experimental results for teams of quadrotors demonstrate online planning for multi-robot exploration and indicate that collision constraints have limited impacts on exploration performance.

Deep Decentralized Multi-task Multi-Agent Reinforcement Learning under Partial Observability

A decentralized single-task learning approach that is robust to concurrent interactions of teammates is introduced, and an approach for distilling single- task policies into a unified policy that performs well across multiple related tasks, without explicit provision of task identity is presented.

The Emergence of Adversarial Communication in Multi-Agent Reinforcement Learning

This work presents a learning model that accommodates individual non-shared rewards and a differentiable communication channel that is common among all agents, and develops a learning algorithm that elicits the emergence of adversarial communications.

Who2com: Collaborative Perception via Learnable Handshake Communication

This paper proposes a multi-stage handshake communication mechanism where the neural network can learn to compress relevant information needed for each stage, and shows that for the semantic segmentation task, the handshake communication method significantly improves accuracy by approximately 20% over decentralized baselines, and is comparable to centralized ones using a quarter of the bandwidth.

Submodular Optimization for Coupled Task Allocation and Intermittent Deployment Problems

In this letter, a formulation for optimizing coupled submodular maximization problems with provable sub-optimality bounds is demonstrated and a greedy algorithm for solving this class of problem is presented, along with sub- optimality guarantees.