A Dynamically Feasible Fast Replanning Strategy with Deep Reinforcement Learning
- PDF / 3,300,586 Bytes
- 17 Pages / 595.224 x 790.955 pts Page_size
- 114 Downloads / 196 Views
A Dynamically Feasible Fast Replanning Strategy with Deep Reinforcement Learning Mehmet Hasanzade1
· Emre Koyuncu2
Received: 3 May 2020 / Accepted: 8 October 2020 © Springer Nature B.V. 2020
Abstract In this work, we aim to develop a fast trajectory replanning methodology enabling highly agile aerial vehicles to navigate in cluttered environments. By focusing on reducing complexity and accelerating the replanning problem under strict dynamical constraints, we employ the b-spline theory with local support property for defining the high dimensional agile flight trajectories. We utilize the differential flatness model of an aerial vehicle, allowing us to directly map the desired output trajectory into input states to track a high dimensional trajectory. Dynamically feasible replanning problem is addressed through regenerating the local b-splines with control point reallocation. As the geometric form of the trajectory based on the location of the control points and the knot intervals, the control point reallocation for fast replanning with dynamical constraints is turned into a constrained optimization problem and solved through deep reinforcement learning. The proposed methodology enables generating dynamically feasible local trajectory segments, which are continuous to the existing, hence provides fast local replanning for collision avoidance. The DRL agent is trained with different environmental complexities, and through the batch simulations, it is shown that the proposed methodology allows to solve fast trajectory replanning problem under given or hard dynamical constraints and provide real-time applicability for such collision avoidance applications in agile unmanned aerial vehicles. Hardware implementation tests of the algorithm with the agile trajectory tracker to a small UAV can bee seen in the following video link: https://youtu.be/8IiLQFQ3V0E. Keywords Agile unmanned aerial vehicles · Trajectory replanning · Deep reinforcement learning
1 Introduction Not long ago, the operations or the applications requiring high-performance guided and navigation would have required the use of tactical-size unmanned aerial vehicles. The main reason for this was that high-performance algorithms required bigger or heavier avionics with high computing capabilities or reliable communication buses linked with the ground systems. However, with the development of technology, these capabilities can now be achieved in smaller avionics, making it possible onboard for small-size unmanned aerial Mehmet Hasanzade
[email protected] 1
Controls and Avionics Research Group, Aerospace Research Center (ITU ARC), Istanbul Technical University, Istanbul, Turkey
2
Department of Aeronautics Engineering, Controls and Avionics Research Group, Aerospace Research Center (ITU ARC), Istanbul Technical University, Istanbul, Turkey
vehicles. New lightweight sensory systems enabled small unmanned systems to have advanced “situational awareness” and allow them to be capable of performing complex missions. Yet, guidance, navigation, and motio
Data Loading...
