Log-Based Reward Field Function for Deep-Q-Learning for Online Mobile Robot Navigation

Path planning is one of the major challenges while designing a mobile robot. In this paper, we implemented Deep-Q-Learning algorithms for autonomous navigation task in wheel mobile robot. We proposed a log-based reward field function to incorporate with Deep-Q-Learning algorithms. The performance of the proposed algorithm is verified in simulated environment and physical environment. Finally, the accuracy of the performance of the obstacle avoidance ability of the robot is measured based on hit rate metrics.

[1]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[2]  Mohammad A. Jaradat,et al.  Reinforcement based mobile robot navigation in dynamic environment , 2011 .

[3]  M. Zacksenhouse,et al.  A neural network architecture for cue-based motion planning , 1988, Proceedings of the 27th IEEE Conference on Decision and Control.

[4]  L Poole David,et al.  Artificial Intelligence: Foundations of Computational Agents , 2010 .

[5]  Dong Yu,et al.  Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[6]  Weria Khaksar,et al.  Automatic navigation of mobile robots in unknown environments , 2013, Neural Computing and Applications.

[7]  Dayal R. Parhi,et al.  Controlling the Motion of an Autonomous Mobile Robot Using Various Techniques: a Review , 2013 .

[8]  Oscar Castillo,et al.  A survey on nature-inspired optimization algorithms with fuzzy logic for dynamic parameter adaptation , 2014, Expert Syst. Appl..

[9]  M. Puterman.  Chapter 8 Markov decision processes , 1990 .

[10]  Henry Hexmoor,et al.  A fuzzy logic approach to robotic path planning with obstacle avoidance , 1986, 1986 25th IEEE Conference on Decision and Control.

[11]  Peter Dayan,et al.  Technical Note: Q-Learning , 2004, Machine Learning.

[12]  Robin De Keyser,et al.  Heuristic approaches in robot path planning: A survey , 2016, Robotics Auton. Syst..

[13]  Alex Graves,et al.  Playing Atari with Deep Reinforcement Learning , 2013, ArXiv.

[14]  Kai Oliver Arras,et al.  Inverse Reinforcement Learning algorithms and features for robot navigation in crowds: An experimental comparison , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[15]  Lotfi A. Zadeh,et al.  Fuzzy Sets , 1996, Inf. Control..