Active Inference in Robotics and Artificial Agents: Survey and Challenges

Abstract

Active inference is a mathematical framework which originated in computational neuroscience as a theory of how the brain implements action, perception and learning. Recently, it has been shown to be a promising approach to the problems of state-estimation and control under uncertainty, as well as a foundation for the construction of goal-driven behaviours in robotics and artificial agents in general. Here, we review the state-of-the-art theory and implementations of active inference for state-estimation, control, planning and learning; describing current achievements with a particular focus on robotics. We showcase relevant experiments that illustrate its potential in terms of adaptation, generalization and robustness. Furthermore, we connect this approach with other frameworks and discuss its expected benefits and challenges: a unified framework with functional biological plausibility using variational Bayesian inference.

arXiv commentThis manuscript is under review in a IEEE journal

Cited by in corpus (8)

• The Free Energy Principle for Perception and Action: A Deep Learning Perspective
• Interactive inference: a multi-agent model of cooperative joint actions
• Modeling motor control in continuous-time Active Inference: a survey
• Geometric Methods for Sampling, Optimisation, Inference and Adaptive Agents
• Observation-Augmented Contextual Multi-Armed Bandits for Robotic Search and Exploration
• Modeling arousal potential of epistemic emotions using Bayesian information gain: Inquiry cycle driven by free energy fluctuations
• Predictive Coding Networks and Inference Learning: Tutorial and Survey
• Relations between entropy rate, entropy production and information geometry in linear stochastic systems