Online reinforcement learning for the shape morphing adaptive control of 4D printed shape memory polymer

Abstract

Combining 3D printing and smart materials, 4D printing technologies enable the printed actuators to further change their shapes or other properties after prototyping. However, the shape morphing of 4D printed actuators suffers from poor controllability and low precision. One of the main challenges is that the 4D printed actuators are hard to be modeled and it is difficult to develop an appropriate controller for them. In this study, various popular reinforcement learning (RL) methods are applied to address the problem of online and adaptive model-free control of 4D printed shape memory polymer (SMP). Their training efficiencies are compared and an adaptive LQR controller based on Q learning is developed to realize efficient online learning. The RL controller achieves precise and quick shape control within 2−−3 learning episodes and is adaptive to the changing properties of SMP. The RL controller performance is then compared with a model-based LQR controller and shows high control precision and excellent adaptability to the varying control plant.