Online Pareto-Optimal Decision-Making for Complex Tasks Using Active Inference

When a robot autonomously performs a complex task, it frequently must balance competing objectives while maintaining safety. This becomes more difficult in uncertain environments with stochastic outcomes. Enhancing transparency in the robot’s behavior and aligning with user preferences are also crucial. This article introduces a novel framework for multiobjective reinforcement learning that ensures safe task execution, optimizes tradeoffs between objectives, and adheres to user preferences. The framework has two main layers: a multiobjective task planner and a high-level selector. The planning layer generates a set of optimal tradeoff plans that guarantee satisfaction of a temporal logic task. The selector uses active inference to decide which generated plan best complies with user preferences and aids learning. Operating iteratively, the framework updates a parameterized learning model based on collected data. Case studies and benchmarks on both manipulation and mobile robots show that our framework outperforms other methods and (i) learns multiple optimal tradeoffs, (ii) adheres to a user preference, and (iii) allows the user to adjust the balance between (i) and (ii).