In this paper a controller design is proposed to get obstacle free trajectories in a three dimensional urban environment for unmanned air vehicles (UAVs). The controller has a two-layer architecture. In the upper layer, vision-inspired Grossberg neural network is proposed to get the shortest distance paths. In the bottom layer, a model predictive control (MPC) based controller is used to obtain dynamically feasible trajectories. Simulation results are presented for to demonstrate the potential of the approach.
V. Yadav et al., "Neural Network Approach for Obstacle Avoidance in 3-D Environments for UAVs," Proceedings of the 2006 American Control Conference, Institute of Electrical and Electronics Engineers (IEEE), Jan 2006.
The definitive version is available at http://dx.doi.org/10.1109/ACC.2006.1657288
2006 American Control Conference
Mechanical and Aerospace Engineering
Keywords and Phrases
Aerospace Control; Collision Avoidance; Control Engineering Computing; Controller Design; Model Predictive Control Based Controller; Neural Nets; Obstacle Avoidance; Obstacle Free Trajectories; Predictive Control; Remotely Operated Vehicles; Unmanned Air Vehicles; Vision-Inspired Grossberg Neural Network
Article - Conference proceedings
© 2006 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.