Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks

Tolstaya, Ekaterina; Paulos, James; Kumar, Vijay; Ribeiro, Alejandro

doi:10.1109/iros51168.2021.9636675

Cited by 43 publications

(38 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The communication graph is G = (V, E) with the node set V as the robots and the edge set E as available communication links, and the graph signal x is the relevant features of robot positions and velocities. We use imitation learning to train the AirGNN by mimicing the optimal centralized controller [7]. The dataset consists of 450 trajectories of 100 time steps, which are split into 400 trajectories for training, 25 for validation, and 25 for testing.…”

Section: B Multi-robot Flockingmentioning

confidence: 99%

“…Graph neural networks (GNNs) are one of the key tools to extract features from networked data [1]- [4], and have found wide applications in wireless communications [5], [6], multiagent coordination [7]- [9] and recommendation systems [10]- [12]. GNNs extend conventional convolutional neural networks (CNNs) to graph structures by employing a multi-layered architecture with each layer comprising a graph filter bank and a pointwise nonlinearity [13].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

AirGNNs: Graph Neural Networks over the Air

Gao¹,

Gündüz²

2023

Preprint

View full text Add to dashboard Cite

Graph neural networks (GNNs) are information processing architectures that model representations from networked data and allow for decentralized implementation through localized communications. Existing GNN architectures often assume ideal communication links, and ignore channel effects, such as fading and noise, leading to performance degradation in real-world implementation. This paper proposes graph neural networks over the air (AirGNNs), a novel GNN architecture that incorporates the communication model into the architecture. The AirGNN modifies the graph convolutional operation that shifts graph signals over random communication graphs to take into account channel fading and noise when aggregating features from neighbors, thus, improving the architecture robustness to channel impairments during testing. We propose a stochastic gradient descent based method to train the AirGNN, and show that the training procedure converges to a stationary solution. Numerical simulations on decentralized source localization and multi-robot flocking corroborate theoretical findings and show superior performance of the AirGNN over wireless communication channels.

show abstract

Section: B Multi-robot Flockingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

AirGNNs: Graph Neural Networks over the Air

Gao¹,

Gündüz²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…where NN e and NN v are multi-layer perceptrons (MLPs). While a Graph Network Block can be used to compose a variety of architectures, for this work, we develop a variant of the Aggregation GNN in which the output of every GN stage is concatenated, and finally, processed by a linear output transform f out [6], [26].…”

Section: A Aggregation Graph Neural Networkmentioning

confidence: 99%

“…Large scale swarms of robots have demonstrated utility in solving many real world problems including rapid environmental mapping [1], [2], target tracking [3], search after natural disasters [4], [5] and exploration [6]. In many of these scenarios, robot teams must operate in harsh environments without existing communication infrastructure, requiring the formation of ad-hoc networks to exchange information.…”

Section: Introductionmentioning

confidence: 99%

Learning Connectivity for Data Distribution in Robot Teams

Tolstaya¹,

Landon²,

Mox³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Many algorithms for control of multi-robot teams operate under the assumption that low-latency, global state information necessary to coordinate agent actions can readily be disseminated among the team. However, in harsh environments with no existing communication infrastructure, robots must form ad-hoc networks, forcing the team to operate in a distributed fashion. To overcome this challenge, we propose a task-agnostic, decentralized, low-latency method for data distribution in ad-hoc networks using Graph Neural Networks (GNN). Our approach enables multi-agent algorithms based on global state information to function by ensuring it is available at each robot. To do this, agents glean information about the topology of the network from packet transmissions and feed it to a GNN running locally which instructs the agent when and where to transmit the latest state information. We train the distributed GNN communication policies via reinforcement learning using the average Age of Information as the reward function and show that it improves training stability compared to task-specific reward functions. Our approach performs favorably compared to industry-standard methods for data distribution such as random flooding and round robin. We also show that the trained policies generalize to larger teams of both static and mobile agents.⋆ Indicates equal contribution.

show abstract

“…They address simplified problems that often exclude common real-world factors such as resource and capacity constraints [23], [25], [26], [31]). 2) They are mostly focused on smaller sized problems (≤ 100 tasks) [30], [32], [34], [35], with their scalability remaining unclear.…”

Section: Introductionmentioning

confidence: 99%

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

Paul¹,

Ghassemi²,

Chowdhury³

2022

Preprint

View full text Add to dashboard Cite

This paper presents a novel graph reinforcement learning (RL) architecture to solve multi-robot task allocation (MRTA) problems that involve tasks with deadlines and workload, and robot constraints such as work capacity. While drawing motivation from recent graph learning methods that learn to solve combinatorial optimization (CO) problems such as multi-Traveling Salesman and Vehicle Routing Problems using RL, this paper seeks to provide better performance (compared to non-learning methods) and important scalability (compared to existing learning architectures) for the stated class of MRTA problems. The proposed neural architecture, called Capsule Attention-based Mechanism or CapAM acts as the policy network, and includes three main components: 1) an encoder: a Capsule Network based node embedding model to represent each task as a learnable feature vector; 2) a decoder: an attention-based model to facilitate a sequential output; and 3) context: that encodes the states of the mission and the robots.To train the CapAM model, the policy-gradient method based on REINFORCE is used. When evaluated over unseen scenarios, CapAM demonstrates better task completion performance and >10 times faster decision-making compared to standard non-learning based online MRTA methods. 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.

show abstract

Multi-Robot Coverage and Exploration using Spatial Graph Neural Networks

Cited by 43 publications

References 21 publications

AirGNNs: Graph Neural Networks over the Air

AirGNNs: Graph Neural Networks over the Air

Learning Connectivity for Data Distribution in Robot Teams

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

Contact Info

Product

Resources

About