Quadcopter control improved by adaptive PID tuned with A3C learning
Design of Adaptive PID Controller Based On Asynchronous Advantage Actor Critic Learning Method for QuadCopter Control
Robotics
Summary
Quadcopters are tricky to control because of their complex movements and sensitivity to disturbances. The authors combined a standard PID controller with a type of reinforcement learning called Asynchronous Advantage Actor-Critic (A3C) to adjust the controller automatically. Their experiments showed this combination leads to better control performance and faster learning than a similar approach without asynchronous agents. This method helps the quadcopter follow its intended path more accurately by improving the underlying control settings.
What this means in practice
- •For drone developers: Enhance quadcopter flight stability by integrating A3C-tuned PID controllers for more precise attitude and trajectory control.
- •For robotics engineers: Use asynchronous reinforcement learning to optimize control parameters dynamically for nonlinear robotic systems similar to quadcopters.
Authors
Ali Jokar, Aria Alasty
Abstract
Quadcopters offer great utility in many applications, but their nonlinear nature and disturbance sensitivity present great control challenges. Basic PID controllers are generally not sophisticated enough to cope with these complexities. This paper suggests a control system that integrates the Asynchronous Advantage Actor-Critic (A3C) algorithm with a PID controller for quadcopter attitude and trajectory tracking. The A3C controller uses parallel agents to optimize PID parameters dynamically using a neural network. A system identification module for the complementary system makes predictions about system states for optimal control policy. The proposed framework was compared with a standard actor-critic (A2C) model. Simulation results verify that they both track accurately. However, the A3C-based controller converges much more for the loss function, as evidenced by reward figures and loss curves, demonstrating better parameter optimization. This shows that A3C-based approach results in improved performance for the control of quadcopter, effectively integrating reinforcement learning and traditional control to achieve higher adaptability.