Adaptive Multi-objective Search in a Swarm vs Swarm Context

As the number of commercial Unmanned Aerial Vehicles (UAVs) or drones continues to increase, there is a need to develop autonomous policing or patrolling mechanism to guard over the swarm of anonymous intruding drones having unknown heterogeneous dynamical behavior and maintain the safety and order in the sky. This paper addresses such a future need and proposes to employ a swarm of patrolling drones that can collectively search a flying zone to find and follow the other anonymous drones effectively for any relevant purposes. Specifically, the key contribution of this research is the development of a robust and flexible search mechanism, which enables the patrolling drones to overcome confusing evasive tactics by the anonymous drones. During the search process, the patrolling drones first scatter to cover a region of interest, and then proceed to find and follow an anonymous drone. For this purpose, a multi-objective optimization problem is formulated. The problem is solved using a few evolutionary algorithms, and performances/results are evaluated through simulation. It is found that multiple particle swarm optimization (MPSO) requires the least processing time.

[1]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[2]  Razvan C. Fetecau,et al.  A nonlocal kinetic model for predator-prey interactions in two dimensions , 2012 .

[3]  Essameddin Badreddin,et al.  Cooperative pursue in pursuit-evasion games with unmanned aerial vehicles , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[4]  Leandro dos Santos Coelho,et al.  Multi-objective grey wolf optimizer: A novel algorithm for multi-criterion optimization , 2016, Expert Syst. Appl..

[5]  Qingfu Zhang,et al.  MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition , 2007, IEEE Transactions on Evolutionary Computation.

[6]  Luo Ding,et al.  Dynamic programming algorithm for the detection of air dim target , 2013 .

[7]  Robin J. Evans,et al.  Peformance of dynamic programming techniques for Track-Before-Detect , 1996 .

[8]  Lee Spector,et al.  Evolving teamwork and coordination with genetic programming , 1996 .

[9]  Graeme D. Ruxton,et al.  Confusion of predators does not rely on specialist coordinated behavior , 2007 .

[10]  Vicenç Gómez,et al.  Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems , 2015, ICAPS.

[11]  S.S. Blackman,et al.  Multiple hypothesis tracking for multiple target tracking , 2004, IEEE Aerospace and Electronic Systems Magazine.

[12]  Mac Schwager,et al.  Eyes in the Sky: Decentralized Control for the Deployment of Robotic Camera Networks , 2011, Proceedings of the IEEE.

[13]  Carlos A. Coello Coello,et al.  Handling multiple objectives with particle swarm optimization , 2004, IEEE Transactions on Evolutionary Computation.

[14]  Markus Olhofer,et al.  A mini-review on preference modeling and articulation in multi-objective optimization: current status and challenges , 2017, Complex & Intelligent Systems.

[15]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[16]  Randal W. Beard,et al.  Cooperative Path Planning for Target Tracking in Urban Environments Using Unmanned Air and Ground Vehicles , 2015, IEEE/ASME Transactions on Mechatronics.