Hybrid System of Reinforcement Learning and Flocking Control in Multi-robot Domain

نویسندگان

  • Ronny Salim Lim
  • Weihua Sheng
  • Hung Manh La
  • Ronny Lim
چکیده

In multi-robot domain, one of the important problems is to achieve cooperation among robots. In this paper we propose a hybrid system that integrates reinforcement learning and flocking control in order to create adaptive and intelligent multi-robot systems. We study two problems of multi-robot concurrent learning of cooperative behaviors: (1) how to generate efficient combination of high level behaviors (discrete states and actions) and low level behavior (continuous states and actions) for multi-robot cooperation; (2) how to coordinate concurrent learning process in a distributed fashion. To evaluate our theoretic framework we apply it to solve the problem of how a multi-robot network learns to avoid predator while maintaining network topology and connectivity. The experiments and simulations are performed to demonstrate the effectiveness of the proposed hybrid system.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An Unsupervised Learning Method for an Attacker Agent in Robot Soccer Competitions Based on the Kohonen Neural Network

RoboCup competition as a great test-bed, has turned to a worldwide popular domains in recent years. The main object of such competitions is to deal with complex behavior of systems whichconsist of multiple autonomous agents. The rich experience of human soccer player can be used as a valuable reference for a robot soccer player. However, because of the differences between real and simulated soc...

متن کامل

Hybrid autonomous control for heterogeneous multi-agent system

Abstmcl-Reinforcement learning is an adaptive and flexible control method for autonomous system. In our previous works, we had proposed a reinforcement learning algorithm for redundant systems: "Q-learning with dynamic structuring of exploration space based on GA (QDSEGAY, and applied it to multi-agent systems. However previous works of the QDSEGA have been restricted to homogeneous agents. In ...

متن کامل

Finite-time Flocking of Multi-agent System with Input Saturation

In this paper a finite-time flocking of multi-agent system with input saturation is investigated. By introducing a new control protocol, it is show that speed and position of agents with double-integrator dynamics converge to the speed and position of leader in finite time. It is assumed that the graph between the agents is connected and that at least one agent is informed of the leader speed a...

متن کامل

Dynamic Obstacle Avoidance by Distributed Algorithm based on Reinforcement Learning (RESEARCH NOTE)

In this paper we focus on the application of reinforcement learning to obstacle avoidance in dynamic Environments in wireless sensor networks. A distributed algorithm based on reinforcement learning is developed for sensor networks to guide mobile robot through the dynamic obstacles. The sensor network models the danger of the area under coverage as obstacles, and has the property of adoption o...

متن کامل

Hybrid Reinforcement Learning and Its Application to Biped Robot Control

A learning system composed of linear control modules, reinforcement learning modules and selection modules (a hybrid reinforcement learning system) is proposed for the fast learning of real-world control problems. The selection modules choose one appropriate control module dependent on the state. This hybrid learning system was applied to the control of a stilt-type biped robot. It learned the ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016