Personalized robotic control via constrained multi-objective reinforcement learning

Reinforcement learning is capable of providing state-of-art performance in end-to-end robotic control tasks. Nevertheless, many real-world control tasks necessitate the balancing of multiple conflicting objectives while simultaneously ensuring that the learned policies adhere to constraints. Additionally, individual users may typically prefer to explore the personalized and diversified robotic control modes via specific preferences. Therefore, this paper presents a novel constrained multi-objective reinforcement learning algorithm for personalized end-to-end robotic control with continuous actions, allowing a trained single model to approximate the Pareto optimal policies for any user-specified preferences. The proposed approach is formulated as a constrained multi-objective Markov decision process, incorporating a nonlinear constraint design to facilitate the agent in learning optimal policies that align with specified user preferences across the entire preference space. Meanwhile, a comprehensive index based on hypervolume and entropy is presented to measure the convergence, diversity and evenness of the learned control policies. The proposed scheme is evaluated on nine multi-objective end-to-end robotic control tasks with continuous action space, and its effectiveness is demonstrated in comparison with the competitive baselines, including classical and state-of-the-art algorithms.

Automated summary

In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.