Effective locomotion in nature happens by transitioning between multiple modes (e.g., walk, run, climb, slither). Despite this, far more mechanistic understanding of terrestrial locomotion has been on how to generate and stabilize around near-steady-state movement in a single mode. Unlike flight and swimming with aero- and hydrodynamics, a major challenge in understanding terrestrial locomotor transitions is the lack of methods to measure and model how animals actively make use of physical interaction with complex terrain. As a result, robots cannot robustly traverse complex terrain, an ability required for critical applications like search and rescue in earthquake rubble and extraterrestrial exploration over Martian rocks. Our BYI work focuses on using an energy landscape approach to elucidate principles of terrestrial locomotor transitions via integration of biomechanics, sensorimotor control, and terrain physics. Recently, we discovered that an energy landscape approach helps understand probabilistic locomotor transitions. In complex terrain, animals’ and robots’ locomotor modes are attracted to basins of a potential energy landscape. We have been developing new methods to enable experiments and modeling of how animals and robots actively cross potential energy barriers and hop between basins to make locomotor transitions. These include: (1) terrain platforms and tracking techniques to quantify animal locomotor transitions and terrain interaction, (2) robots with animal-like motion as physical models, and (3) force sensors to measure physical interaction and understand provide feedback control. I will highlight two recent discoveries enabled by these tools: to make locomotor transitions and traverse complex terrain, (1) legged insects actively adjust head, body, and appendages to lower potential energy barriers; (2) besides vision, limbless snakes must use mechanical sensing to adjust body form and control terrain contact to maintain propulsion and stability.