Abstract | ||
---|---|---|
This paper presents a new path planning method for mobile robots in unknown environments. The structure of the proposed algorithm is a hybrid fuzzy logic neural networks, and hence it benefits from the potentials of these two techniques. For modeling the mobile robot, the proposed system adopts the Braitenberg's automata models that were developed for agents. Wheels of the robot are represented by a bio-inspired neuron of a neural network, where each wheel receives different sensor inputs indicating different signals from either excitatory or inhibitory synapses. Training of the neural network weighting is automatically achieved through the fuzzy system that is developed to adjust the weighting between each synapse and neuron of the network. To assess the performance of the developed algorithm, simulation results are presented. It was shown that the proposed method can successfully navigate the robot to the target, and turn the robot at corners for given desired angles. The methodology proposed herein improves the Braitenberg navigation scheme and offers insights into using biologically inspired systems for path planning. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/MFI.2012.6343040 | MFI |
Field | DocType | ISBN |
Motion planning,Computer science,Fuzzy logic,Artificial intelligence,Mobile robot navigation,Fuzzy control system,Robot,Artificial neural network,Machine learning,Mobile robot,Braitenberg vehicle | Conference | 978-1-4673-2511-0 |
Citations | PageRank | References |
4 | 0.44 | 8 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lei Wang | 1 | 4 | 0.44 |
Simon X. Yang | 2 | 1029 | 124.34 |
Mohammad Biglarbegian | 3 | 225 | 9.57 |