A Two-Stage Reinforcement Learning Approach for Multi-UAV Collision Avoidance Under Imperfect Sensing.
Dawei WangTingxiang FanTao HanJia PanPublished in: IEEE Robotics Autom. Lett. (2020)
Keyphrases
- collision avoidance
- path planning
- reinforcement learning
- dynamic environments
- mobile robot
- visual navigation
- path finding
- motion planning
- state space
- obstacle avoidance
- formation control
- real time
- robot motion
- unmanned aerial vehicles
- fuzzy neural network
- control algorithm
- expert systems
- search algorithm
- multi robot
- feed forward
- artificial neural networks