Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks

@article{Tolstaya2021MultiRobotCA,
  title={Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks},
  author={Ekaterina V. Tolstaya and James Paulos and Vijay R. Kumar and Alejandro Ribeiro},
  journal={2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2021},
  pages={8944-8950}
}
The multi-robot coverage problem is an essential building block for systems that perform tasks like inspection, exploration, or search and rescue. We discretize the coverage problem to induce a spatial graph of locations and represent robots as nodes in the graph. Then, we train a Graph Neural Network controller that leverages the spatial equivariance of the task to imitate an expert open-loop routing solution. This approach generalizes well to much larger maps and larger teams that are… 

Figures and Tables from this paper

Graph Neural Networks in IoT: A Survey

TLDR
A comprehensive review of recent advances in the application of Graph Neural Networks to the IoT field is presented, including a deep dive analysis of GNN design in various IoT sensing environments, an overarching list of public data and source code from the collected publications, and future research directions.

Graph Neural Networks for Multi-Robot Active Information Acquisition

—This paper addresses the Multi-Robot Active In- formation Acquisition (AIA) problem, where a team of mobile robots, communicating through an underlying graph, estimates a hidden state expressing a

LEMURS: Learning Distributed Multi-Robot Interactions

TLDR
It is demonstrated that LEMURS can learn interactions and cooperative behaviors from demonstrations of multi-agent navigation and flocking tasks, and is distributed by construction, enabling the learned control policies to be deployed in robot teams of different sizes.

D2CoPlan: A Differentiable Decentralized Planner for Multi-Robot Coverage

TLDR
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.

Real-Time Survivor Detection System in SaR Missions Using Robots

TLDR
The usage of robots will be introduced for the initial exploration of the calamity site and identify the location of human survivors by examining the video feed captured by them and stream the detected location of the survivor to a centralized cloud server.

Learning Scalable Policies over Graphs for Multi-Robot Task Allocation using Capsule Attention Networks

TLDR
CapAM's advantage in generalizability, and scalability to test problems of size larger than those used in training, are also successfully demonstrated in comparison to a popular approach for learning to solve CO problems, namely the purely attention mechanism.

A Self-Organizing Area Coverage Method for Swarm Robots Based on Gradient and Grouping

TLDR
A coverage method based on gradient and grouping (GGC) is proposed that is suitable for extremely simple robots that lack computing or storage power and can be applied to the submillimeter swarm robot system, which will lay the foundation for micro medicine.

ModGNN: Expert Policy Approximation in Multi-Agent Systems with a Modular Graph Neural Network Architecture

TLDR
This paper introduces ModGNN, a decentralized framework which serves as a generalization of GCNs, providing more flexibility, and performs an ablation analysis to show that the most important component of the framework is one that does not exist in a GCN.

Learning Connectivity for Data Distribution in Robot Teams

TLDR
This work proposes a task-agnostic, decentralized, low-latency method for data distribution in ad-hoc networks using Graph Neural Networks (GNN), and trains the distributed GNN communication policies via reinforcement learning using the average Age of Information as the reward function and shows that it improves training stability compared to task-specific reward functions.

References

SHOWING 1-10 OF 29 REFERENCES

Relational inductive biases, deep learning, and graph networks

TLDR
It is argued that combinatorial generalization must be a top priority for AI to achieve human-like abilities, and that structured representations and computations are key to realizing this objective.

A survey on coverage path planning for robotics

Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks

TLDR
This work learns a single common local controller which exploits information from distant teammates using only local communication interchanges and applies this approach to a decentralized linear quadratic regulator problem and observes how faster communication rates and smaller network degree increase the value of multi-hop information.

Learning Connectivity for Data Distribution in Robot Teams

TLDR
This work proposes a task-agnostic, decentralized, low-latency method for data distribution in ad-hoc networks using Graph Neural Networks (GNN), and trains the distributed GNN communication policies via reinforcement learning using the average Age of Information as the reward function and shows that it improves training stability compared to task-specific reward functions.

Multi-Agent Routing Value Iteration Network

TLDR
A graph neural network based model is proposed that is able to perform multi-agent routing based on learned value iteration in a sparsely connected graph with dynamically changing traffic conditions and significantly outperforms traditional solvers both in terms of total cost and runtime.

On Tools

Graph Policy Gradients for Large Scale Robot Control

TLDR
This paper proposes a new algorithm called Graph Policy Gradients (GPG) that exploits the underlying graph symmetry among the robots that is able to scale better than existing reinforcement methods that employ fully connected networks.

An Efficient Graph Convolutional Network Technique for the Travelling Salesman Problem

This paper introduces a new learning-based approach for approximately solving the Travelling Salesman Problem on 2D Euclidean graphs. We use deep Graph Convolutional Networks to build efficient TSP

Aggregation Graph Neural Networks

TLDR
This work proposes a multi-node aggregation GNN that constructs regional features that are later aggregated into more global features and so on, and shows superior performance in a source localization problem on synthetic graphs and on the authorship attribution problem.

Learning to Plan via Neural Exploration-Exploitation Trees

TLDR
A meta path planning algorithm which can utilize prior experience to drastically reduce the sample requirement for solving new path planning problems, and which can complete the planning tasks with very small search trees and significantly outperforms previous state-of-the-arts on several benchmark problems.