Online Pareto-Optimal Decision-Making for Complex Tasks using Active Inference
This addresses the challenge of safe and transparent decision-making for autonomous robots in complex tasks, though it appears incremental as it builds on existing multi-objective and active inference methods.
The paper tackles the problem of robots balancing competing objectives and user preferences in uncertain environments by introducing a multi-objective reinforcement learning framework with a planner and selector, resulting in improved performance in manipulation and mobile robot benchmarks.
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 paper introduces a novel framework for multi-objective reinforcement learning that ensures safe task execution, optimizes trade-offs between objectives, and adheres to user preferences. The framework has two main layers: a multi-objective task planner and a high-level selector. The planning layer generates a set of optimal trade-off 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 trade-offs, (ii) adheres to a user preference, and (iii) allows the user to adjust the balance between (i) and (ii).