Recent Trends in Insect and Robot Navigation through the Lens of Reinforcement Learning

Bees are among the master navigators of the insect world. Despite impressive advances in robot navigation research, the performance of these insects is still unrivaled by any artificial system in terms of training efficiency and generalization capabilities, particularly considering the limited computational capacity. On the other hand, computational principles underlying these extraordinary feats are still only partially understood. The theoretical framework of reinforcement learning (RL) provides an ideal focal point to bring the two fields together for mutual benefit. While RL has long been at the core of robot navigation research, current computational theories of insect navigation are not commonly formulated within this framework, but largely as an associative learning process implemented in the insect brain, especially in a region called the mushroom body. Here we argue that this neural substrate can be understood as implementing a certain class of relatively simple RL algorithms, capable of integrating distinct components of a navigation task, reminiscent of hierarchical RL models used in robot navigation. The efficiency of insect navigation is likely rooted in an efficient and robust internal representation of space, linking retinotopic (egocentric) visual input with the geometry of the environment. We discuss how current models of insect and robot navigation are exploring representations beyond classical, complete map-like representations, with spatial information being embedded in the respective latent representations to varying degrees.