Despite decades of research and significant investment, robotics engineers have yet to develop a robot that can match the walking or running abilities of animals. Many animals exhibit remarkable feats that current robots cannot replicate, such as the wildebeest’s endurance in long migrations, the mountain goat’s ability to scale cliffs, and the cockroach’s resilience even after losing a limb.
Dr. Max Donelan, a Professor at Simon Fraser University, highlights this disparity, emphasizing that robots lack the endurance, agility, and robustness seen in animals. To dissect the reasons behind this gap, an interdisciplinary team of scientists and engineers conducted a comprehensive study comparing running robots with their biological counterparts, culminating in a paper published in Science Robotics.
The team, which included Drs. Sam Burden, Tom Libby, Kaushik Jayaram, and Simon Sponberg, focused on five key subsystems—Power, Frame, Actuation, Sensing, and Control—present in both robots and animals. Surprisingly, the study found that, in terms of engineering metrics, fabricated components often outperformed biological ones. However, animals excelled in integrating and controlling these components to achieve remarkable movement capabilities.
While acknowledging the rapid progress in robotics, especially considering the relatively short development time compared to biological evolution, the researchers underscored the need for further advancements in integrating and controlling robotic hardware. They emphasized the potential applications of running robots in various fields, from logistics to hazardous environment exploration.
The study aims to guide future developments in robotics, focusing on enhancing integration and control rather than solely improving hardware. By learning from biology’s integration principles, researchers aspire to make running robots as efficient, agile, and robust as their biological counterparts.
