CART terrain-adaptation paper evaluates context-aware locomotion on Spot

Evidence notes

• The CART paper introduces context-aware terrain adaptation using temporal sequence selection for legged robots.
• The authors evaluate the method in simulation with Unitree Go2 and ANYmal-C and in real-world experiments with Boston Dynamics Spot.
• The results report improved locomotion behavior on varied terrain by linking exteroceptive terrain appearance with proprioceptive physical interaction.