In an industrial environment, modular robots aren’t hot news. But what if those modular robots were able to autonomously make changes to themselves? A Cornell Engineering-led research team recently asked that question and developed a prototype amorphous robot that can change its shape to accomplish various tasks.
A recent Cornell Chronicle article by Melanie Lefkowitz explains that the robot is made up of a sensor module featuring several cameras and a computer for collecting the processing data from its surroundings. Software includes a high-level planner, which directs actions and reconfiguration, and perception algorithms to map, navigate and classify surroundings.
Connected to the main sensor module are wheeled, cube-shaped modules, which detach and reattach using magnets to change the overall shape of the robot. The cubes use Wi-Fi to communicate with one another and the sensor module, and were developed by researchers at the University of Pennsylvania.
“This is the first time modular robots have been demonstrated with autonomous reconfiguration and behavior that is perception-driven,” Hadas Kress-Gazit, associate professor in the Sibley School of Mechanical and Aerospace Engineering and principal investigator on the project, said in the article. “We are creating a modular system that is able to do different tasks autonomously. By changing the high-level task, it totally changes its behavior.”
The prototype bot can form into 57 possible configurations and perform 97 different behaviors. When the robot receives a task, the planning software searches through a library of configurations and behaviors to choose those that best fit the assignment.
“I want to tell the robot what it should be doing, what its goals are, but not how it should be doing it,” Kress-Gazit said in the article. “I don’t actually prescribe, ‘Move to the left, change your shape.’ All these decisions are made autonomously by the robot.”
As you can see in the video below, the team conducted three experiments to test the bot’s effectiveness. From those experiments, the researchers found some error in the hardware and low-level software of the robot. Further improvements in these areas would allow autonomous modular robots to benefit numerous industries.
“Modular robots in general are just fascinating systems, because you’re not restricted by one shape, so there’s a lot of flexibility,” Kress-Gazit said in the article. “The hardware is still in research stages, but if we had commercial modular robots, they would be very useful for anything where the environment changes significantly and the robot should adapt to its environment as well.”
The research was funded by the National Science Foundation, and recently published in Science Robotics.