Studying Cheetahs Through the Lens of Robotics

In his presentation, Amir Patel from the African Robotics Unit (ARU) at the University of Cape Town elaborates how his research team drew inspiration from the biomechanics of cheetahs—which are known for their extraordinary agility, rapid acceleration, and ability to execute sharp turns—to develop more agile robots. The ARU worked with three cheetah centers (Cheetah Outreach, Cheetah Conservation & Research Centre, and Ashia) in the area where the cheetahs are accustomed to people.

The talk delves into the various methodologies employed—including multibody modeling, feedback control, trajectory optimization, state estimation, and deep learning—to analyze and translate the biomechanics of cheetahs into robotic control strategies. Amir discusses the challenges associated with capturing wildlife biomechanics using GPS-IMU collars, extensive camera setups, and unmanned aerial vehicles. These difficulties prompted the ARU team to create WildPose, a long-range motion capture system that employs lidar and telescopic lenses to study cheetahs in their natural habitat. Furthermore, the presentation highlights the team's bipedal robot named ""Baleka"" (Xhosa word meaning “to sprint”) and a quadrupedal robot prototype called ""Kemba"" (Zulu “inkemba,” or sword ) demonstrating impressive agility. It also covers the use of MATLAB^®, Simulink^®, Simscape™, and Simulink Real-Time™ by the research team to model, simulate, and control the robots.