Learning of Long-Horizon Sparse-Reward Robotic Manipulator Tasks with Base Controllers
This work aims to improve the sample efficiency and performance of DRL for long-horizon sparse-reward robotic manipulation tasks, which is a significant problem for robotics researchers and practitioners.
This paper addresses the challenge of long-horizon sparse-reward robotic manipulation tasks by integrating existing traditional base controllers into a Deep Reinforcement Learning (DRL) framework. The proposed method, built on DDPG, incorporates base controllers into exploration, value learning, and policy updates, and also synthesizes multiple controllers. Experiments show that the learned policies consistently outperform base controllers and achieve orders of magnitude improvement in sample efficiency compared to previous learning-from-demonstrations methods.
Deep Reinforcement Learning (DRL) enables robots to perform some intelligent tasks end-to-end. However, there are still many challenges for long-horizon sparse-reward robotic manipulator tasks. On the one hand, a sparse-reward setting causes exploration inefficient. On the other hand, exploration using physical robots is of high cost and unsafe. In this paper, we propose a method of learning long-horizon sparse-reward tasks utilizing one or more existing traditional controllers named base controllers in this paper. Built upon Deep Deterministic Policy Gradients (DDPG), our algorithm incorporates the existing base controllers into stages of exploration, value learning, and policy update. Furthermore, we present a straightforward way of synthesizing different base controllers to integrate their strengths. Through experiments ranging from stacking blocks to cups, it is demonstrated that the learned state-based or image-based policies steadily outperform base controllers. Compared to previous works of learning from demonstrations, our method improves sample efficiency by orders of magnitude and improves the performance. Overall, our method bears the potential of leveraging existing industrial robot manipulation systems to build more flexible and intelligent controllers.