Perception of dynamics on rigid slopes via improved trajectory and compliant foot-slope interaction for a hexapod robot

The invalidation of hexapod robots' climbing trajectory decisions is caused by slope perception deviations, triggering joint internal force antagonism, foot slippage, and even robot instability or overturning. Thus, slope climbing remains a critical challenge for hexapod robots. This study proposes an optimized cycloidal composite trajectory tailored for flat-to-slope transitions. Leveraging stress-distributed perception of compliant feet as a core input, we capture real-time foot-slope contact forces and joint bending moment pulses; inverse derivation of pose-driving time-domain information throughout the full foot-placement cycle enables an identification with relatively high precision—a valid approach reducing reliance on complex vision-based sensing systems. Specifically, we analyze the smooth transition from flat ground to sloped terrain and address kinematic/dynamic modeling under limited mechanical workspace and actuation constraints. Forward/inverse kinematics and dynamics of swing legs during the detection phase are solved numerically. The proposed framework emphasizes dynamic model-based decision-making and trajectory smoothing using planar stress distribution and force perception instead of vision, enabling intelligent prediction of sloped environments, particularly in transition zones between flat terrain and slopes.