Optimal path-planning for mobile robots to find a hidden target in an unknown environment based on machine learning

Using mobile robots in disaster areas can reduce risks and the search time in urban search and rescue operations. Optimal path-planning for mobile robotics can play a key role in the reduction of the search time for rescuing victims. In order to minimize the search time, the shortest path to the target should be determined. In this paper, a new integrated Reinforcement Learning—based method is proposed to search and find a hidden target in an unknown environment in the minimum time. The proposed algorithm is developed in two main phases. Depending on whether or not the mobile robot receives the signal from the hidden target, phases I or II of the proposed algorithm can be carried out. Then, the proposed algorithm is implemented on an e-puck robot in an urban environment which is simulated within Webots software. Finally, to demonstrate the efficiency of the proposed method and to verify it, the computational results from the proposed method are compared with three conventional methods from the literature.

[1]  Iván Santos-González,et al.  Secure system based on UAV and BLE for improving SAR missions , 2020, J. Ambient Intell. Humaniz. Comput..

[2]  Lisa Lee.  Robotic Search & Rescue via Online Multi-task Reinforcement Learning , 2015, ArXiv.

[3]  Jan Faigl,et al.  Multi-Goal Path Planning Using Self-Organizing Map with Navigation Functions , 2011, ESANN.

[4]  Richard L. Church,et al.  Finding shortest paths on real road networks: the case for A* , 2009, Int. J. Geogr. Inf. Sci..

[5]  Domenico Pascarella,et al.  Parallel and distributed computing for UAVs trajectory planning , 2015, J. Ambient Intell. Humaniz. Comput..

[6]  N. Geng,et al.  PSO-Based Robot Path Planning for Multisurvivor Rescue in Limited Survival Time , 2014 .

[7]  Robin R. Murphy,et al.  Human-robot interactions during the robot-assisted urban search and rescue response at the World Trade Center , 2003, IEEE Trans. Syst. Man Cybern. Part B.

[8]  Samarjit Kar,et al.  Multi-criteria shortest path for rough graph , 2018, J. Ambient Intell. Humaniz. Comput..

[9]  Ronald L. Rivest,et al.  Introduction to Algorithms , 1990 .

[10]  Jianhua Wu,et al.  Dynamic Obstacle Avoidance for an Omnidirectional Mobile Robot , 2010, J. Robotics.

[11]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[12]  Kurt Mehlhorn,et al.  Review of algorithms and data structures: the basic toolbox by Kurt Mehlhorn and Peter Sanders , 2011, SIGA.

[13]  Ariel Felner.  Position Paper: Dijkstra's Algorithm versus Uniform Cost Search or a Case Against Dijkstra's Algorithm , 2011, SOCS.

[14]  Francesco Mondada,et al.  The e-puck, a Robot Designed for Education in Engineering , 2009 .

[15]  Saiedeh Gholami,et al.  A new effective algorithm for on-line robot motion planning , 2014 .

[16]  Brian Yamauchi,et al.  Frontier-based exploration using multiple robots , 1998, AGENTS '98.

[17]  Mahdi Fakoor,et al.  Revision on fuzzy artificial potential field for humanoid robot path planning in unknown environment , 2015 .