An underwater hexapod robot, driven by six C-shaped legs and eight thrusters, has the potential to traverse diverse terrains with unknown deformable properties, which can lead to unknown leg–terrain interaction forces. However, it is hard to use exteroceptive sensors such as cameras and sonars to recognize these properties. Here we propose a method to perceive the interaction forces and feed them into a controller for determining thrust inputs. The key idea lies in using supervised learning to obtain the properties from reliable proprioceptive sensory data. First, we propose a new expression called zero moment point (ZMP) bias that can indirectly represent the leg–terrain interaction force, removing the effects caused by gravity, buoyancy, and thrust. Second, we gather a walking cycle’s discrete ZMP biases and then parameterize them as polynomials. Third, we use several previous walking cycles’ parameterized biases to predict the current walking cycle’s biases to generate the needed pitch and roll moments. Finally, we propose a terrain-adaptive locomotion controller for the robot, which incorporates these moments into a base control module and uses thrust to compensate for the interaction force for smooth walking. Extensive indoor pool and wild lake hardware experiments confirm our method’s effectiveness.
