Individual distinguishing pheromone in a multi-robot system for a Balanced Partitioned Surveillance task

A multi-robot system conceived for accomplishing surveillance tasks is designed. The surveillance task addressed differs from that usually considered in the sense that additional requirements must be satisfied, namely: the environment must be virtually partitioned into so many disjoint areas of equal task cost as the number of robots; and each robot must execute the surveillance task restricted to only one area. The strategy adopted to coordinate the robots is based on a stigmergic communication mechanism based on the following property: every robot releases its own pheromone (individual distinguishing pheromone), that is, every robot is able to distinguish the pheromone it releases from that pheromone any other robot does. The multi-robot system operates in a totally decentralized, reactive and on-line mode. Different environment structures and number of robots are considered to build experiments. The respective results confirm the capabilities of the multi-robot system proposed. Specifically, as the robots execute the task, it is possible to observe some behaviors emerge: firstly, dispersion; after that, navigation confinement. In the end, the system dynamics correspond to the task execution according to the desired requirements, at least, approximately: the environment is totally and unceasingly monitored and virtually partitioned into balanced disjoint areas each of which surveilled by only one robot.

[1]  Nicholas R. Jennings,et al.  Near-optimal continuous patrolling with teams of mobile information gathering agents , 2013, Artif. Intell..

[2]  David A. Anisi,et al.  Cooperative Minimum Time Surveillance With Multiple Ground Vehicles , 2010, IEEE Transactions on Automatic Control.

[3]  Marco Wiering,et al.  2011 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) , 2011, IJCNN 2011.

[4]  Habiba Drias,et al.  A multi-agent approach for integrated emergency vehicle dispatching and covering problem , 2012, Eng. Appl. Artif. Intell..

[5]  W. Xiang,et al.  Ant colony intelligence in multi-agent dynamic manufacturing scheduling , 2008, Eng. Appl. Artif. Intell..

[6]  Madan Lal Mittal,et al.  A multi-agent system for decentralized multi-project scheduling with resource transfers , 2013 .

[7]  Deepa Krishnan A Distributed Self-Adaptive Intrusion Detection System for Mobile Ad-hoc Networks Using Tamper Evident Mobile Agents☆ , 2015 .

[8]  Valentin Robu,et al.  A multi-agent platform for auction-based allocation of loads in transportation logistics , 2008, Expert Syst. Appl..

[9]  Erion Plaku,et al.  Reactive Motion Planning for Unmanned Aerial Surveillance of Risk-Sensitive Areas , 2015, IEEE Transactions on Automation Science and Engineering.

[10]  Beom Hee Lee,et al.  Faulty robot rescue by multi-robot cooperation , 2013, Robotica.

[11]  Keng Peng Tee,et al.  Multi-Robot Item Delivery and Foraging: Two Sides of a Coin , 2015, Robotics.

[12]  David Portugal,et al.  Distributed multi-robot patrol: A scalable and fault-tolerant framework , 2013, Robotics Auton. Syst..

[13]  Qi Hao,et al.  A Multi-Agent-Based Intelligent Sensor and Actuator Network Design for Smart House and Home Automation , 2013, J. Sens. Actuator Networks.

[14]  Alan K. Mackworth,et al.  Multi-robot repeated area coverage , 2013, Auton. Robots.

[15]  Ademir Aparecido Constantino,et al.  A multi-pheromone stigmergic distributed robot coordination strategy for fast surveillance task execution in unknown environments , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).