Many animals establish, learn and optimise routes between locations to commute efficiently. Bees and ants commute between their nest and locations with resources, such as flowers or cookies provided by the experimenters, along habitual paths. Routes, in nature, can span several hundred meters to kilometers. While almost all knowledge on insect route following have been studied in the natural habitat, little is known to the underlying mechanisms of route learning and following. Indeed, the longer the route is, the harder is the systematic investigation of route establishment, learning and following. Therefore, we developed behavioural experimental assays to investigate route formation and following in indoor and controlled environments. Our assays rely on a range of techniques to investigate the mechanisms underlying route formation, from skinnerian training to unsupervised route formation. We follow individually marked bumblebees in a variety of terrains from simple routes, a couple or left and right maneuvers, to routes in highly cluttered environments. We record the bee’s behavior using high-frequency video footage by combining online recording techniques and machine learning supervised data-analyses. Finally, we develop methods to enhance our understanding of route following by insects, from clustering techniques to identify routes from paths, to closed-loop driven agent serving as hypothetical neuronal correlate of the bee’s behaviour.