Multi-agent Robotic Systems in Collaborative Robotics

Vorotnikov, Sergey; Ermishin, Konstantin; Nazarova, Anaid V.; Yuschenko, Arkady

doi:10.1007/978-3-319-99582-3_28

Cited by 49 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The strategy of group control and management directly affects the composition and the structure of the CRS participants [6][7][8]. As a whole, group management strategies are divided into centralized and decentralized, depending on the operational characteristics of the participants of the collaborative robotic system.…”

Section: Principles Of Crs Formationmentioning

confidence: 99%

Analysis of the Allocation and Implementation of Tasks in the Heterogeneous Team of the Collaborative Robotic System

Galin

Mamchenko

Meshcheryakov

2021

Electromechanics and Robotics

View full text Add to dashboard Cite

The paper discusses the functioning of a collaborative robotic system that includes both collaborative robots (cobots) and humans (operators). Group control (management) methods and algorithms to solve the problem of increasing the efficiency of human-robot interaction are proposed to be used. The article deals with the principles of the formation and functioning of heterogeneous teams within a collaborative robotic system (CRS), as well as the methods of collective tasks allocation among the participants of these teams. The members of the CRS heterogeneous team perform jointly their specific sub-tasks in order to achieve a single global goal of the system (or accomplish its single common mission). The methods and algorithms of task allocation for the CRS members are considered in the context of a hybrid management strategy. The authors also propose a structural scheme of the modular control system of the CRS (based on a distributed hierarchical principle) that takes into account the indeterminacy of the external environment and the lack of posterior information provided.

show abstract

Section: Principles Of Crs Formationmentioning

confidence: 99%

Analysis of the Allocation and Implementation of Tasks in the Heterogeneous Team of the Collaborative Robotic System

Galin

Mamchenko

Meshcheryakov

2021

Electromechanics and Robotics

View full text Add to dashboard Cite

show abstract

“…An architecture, consisting of the agent-executor and the agent-user, is used to demonstrate method of automated usability evaluation process (Vorotnikov et al, 2018). The agent-executor (AE) has two channels.…”

Section: Agent Architecture For Automation Usability Evaluationmentioning

confidence: 99%

Automating Usability Evaluation of Visual User Interfaces in the Tile Logic

Devyatkov¹,

Tipsin²

2019

International Conferences Interfaces and Human Computer Interaction 2019; Game and Entertainment Technologies 2019; And Compute

View full text Add to dashboard Cite

This article presents the methodology of formal automated evaluation of user interface usability. The tiles logic was selected as the language for solving problems in formal usability verification. The paper contains an overview of the related works, explains the reasons for using the tile logic, and considers the principles of verification (proof) of usability as validation of agent's interaction. The differences and advantages of the developed approach from the well-known are shown. A description of the agents interaction in the tile logic is used to automate usability evaluation. The methodology of formal automated usability evaluation is demonstrated by describing the interaction of two simple agents in the tile logic. In conclusion, we present future of use and development of the proposed approach.

show abstract

“…By taking advantage of the deep learning technique, cooperative multi-agent reinforcement learning (MARL) shows great scalability and excellent performance on challenging tasks (Vorotnikov et al, 2018;Wu et al, 2020) such as StarCraft unit micromanagement . As an efficient paradigm of cooperative MARL, centralized training with decentralized execution (CTDE) (Oliehoek et al, 2008;Foerster et al, 2016;Lowe et al, 2017) gains growing attention.…”

Section: Introductionmentioning

confidence: 99%

Greedy-based Value Representation for Optimal Coordination in Multi-agent Reinforcement Learning

Wan¹,

Liu²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

Due to the representation limitation of the joint Q value function, multi-agent reinforcement learning (MARL) methods with linear or monotonic value decomposition suffer from the relative overgeneralization. As a result, they can not ensure the optimal coordination. Existing methods address the relative overgeneralization by achieving complete expressiveness or learning a bias, which is insufficient to solve the problem. In this paper, we propose the optimal consistency, a criterion to evaluate the optimality of coordination. To achieve the optimal consistency, we introduce the True-Global-Max (TGM) principle for linear and monotonic value decomposition, where the TGM principle can be ensured when the optimal stable point is the unique stable point. Therefore, we propose the greedy-based value representation (GVR) to ensure the optimal stable point via inferior target shaping and eliminate the non-optimal stable points via superior experience replay. Theoretical proofs and empirical results demonstrate that our method can ensure the optimal consistency under sufficient exploration. In experiments on various benchmarks, GVR significantly outperforms state-of-the-art baselines.

show abstract

Multi-agent Robotic Systems in Collaborative Robotics

Cited by 49 publications

References 2 publications

Analysis of the Allocation and Implementation of Tasks in the Heterogeneous Team of the Collaborative Robotic System

Analysis of the Allocation and Implementation of Tasks in the Heterogeneous Team of the Collaborative Robotic System

Automating Usability Evaluation of Visual User Interfaces in the Tile Logic

Greedy-based Value Representation for Optimal Coordination in Multi-agent Reinforcement Learning

Contact Info

Product

Resources

About