Hierarchical gaussian reinforcement learning for path planning in uncertain environments
Most of the issues in planning and controlling of robots are caused by uncertainties in the actuators and sensors of robots. Path planning is of paramount importance for autonomous mobile robots. This paper presents a path planning approach that is based on hierarchical Gaussian reinforcement learni...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Research India Publications
2015
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/43050/ http://irep.iium.edu.my/43050/ http://irep.iium.edu.my/43050/1/2_35120-__Paper_code_-_IJAER_ok_20029-20040.pdf |
| Summary: | Most of the issues in planning and controlling of robots are caused by uncertainties in the actuators and sensors of robots. Path planning is of paramount importance for autonomous mobile robots. This paper presents a path planning approach that is based on hierarchical Gaussian reinforcement learning. This approach differs from traditional Q-leaning in two ways: its ability to deal with continuous states and actions and its ability to work in uncertain (nondeterministic) environments. We propose a path planning algorithm for robots in uncertain environments by using hierarchical Gaussian Q-learning. We used Matlab to perform experiments in simulation. The simulation experimental results seem suggest the efficiency of the proposed algorithm in finding optimal paths of autonomous agents. |
|---|