Researchers at the Biohybrid and Organic Robotics Group, led by Vickie Webster-Wood, have developed a groundbreaking model that integrates living muscle tissue into robotic systems. This innovation aims to enable robots to exhibit behaviors typically associated with biological organisms. Unlike conventional actuators, muscle-based actuators can adapt and strengthen through use, allowing robots powered by living tissues not only to move but also to exercise, enhancing their ability to adjust to various environments and perform tasks more efficiently over time.
To harness the potential of biohybrid robots for specific applications, the team focused on understanding how to design and control robots that become stronger with use. Their approach employs reinforcement learning to control a model of a biohybrid robot, adapting as the muscular actuators improve with each attempt at a designated task.
Testing the Adaptive Robot Model
In a recent experiment, the researchers created a soft, worm-like robot composed of 42 living muscles. This robot was tasked with moving toward eight distinct targets within a simulated environment, requiring it to coordinate its muscle contractions differently for each target. To assess the impact of muscle adaptability on the learning process, the team conducted simulations with both static muscles and those that strengthened with use.
“We initially questioned whether the AI agent would be negatively impacted by muscle adaptability,” stated Vickie Webster-Wood, who serves as an associate professor of mechanical engineering. The results revealed a surprising outcome: the adaptable actuators did not hinder the learning process at all.
The team successfully demonstrated that the robot could learn to move toward the eight targets by coordinating its muscle contractions, even as the muscles evolved over time. This adaptability not only facilitated faster learning but also enhanced the robot’s overall performance.
Implications for Future Biohybrid Robotics
These findings mark a significant step toward the development of biohybrid robots capable of adapting to their surroundings in ways similar to animals. “This brings us one step closer to designing and eventually building biohybrid robots that can adapt to the world around them, just like animals do,” Webster-Wood emphasized.
As research progresses, the implications for various fields, including healthcare, manufacturing, and exploration, are profound. Biohybrid robots could potentially lead to advancements in prosthetics, rehabilitation devices, and even autonomous systems that mimic the adaptability found in nature. The ongoing research by the Biohybrid and Organic Robotics Group promises to open new avenues in robotics, bridging the gap between biological systems and engineered machines.
