Designing and Developing a Neural Network Path Planner System on Mekanum Wheel Mobile Robot Platform
The objective of this thesis work is to develop a mekanum wheel mobile robot platform that has a neural network approach in generating the shortest path while avoiding collisions.A mekanum wheel mobile robot platform was designed during Mechatronics system Design 2 project in 7th semester. a Planning, Execution, and Monitoring (PEM) architecture is applied to the system hierarchy to divide the processes into several smaller modules. the monitoring module keeps track of the robot movement. the planner module generates a trajectory to reach the destination point using Artificial Neural Network (ANN) architecture. the proposed ANN is a three layer ΓÇÜ├ä├╢?├æ??100-10-100ΓÇÜ├ä├╢?├æ?? network which consists of an input, a hidden, and an output layer. This ANN model is trained through a supervised learning method using back-propagation algorithm. the network is continuously trained throughout the thesis work. the path is converted into a set of command sequences in data packets which then will be sent to the execution module to be executed.The framework is developed using Qt Creator under C++ programming language and Arduino IDE.
|B01554||(Rack Thesis)||Available but not for loan - Missing|
No other version available