Snake-Like Intelligent Robot in Development

Jie Fu, left, and Cagdas Onal hold an earlier snake-like soft robot (Worcester Polytechnic Institute)

19 October 2017. An engineering team is creating a self-powered robot that can travel independently like a snake through uneven and cluttered surfaces, such as disaster sites. Worcester Polytechnic Institute in Massachusetts recently received a $400,000 award from National Science Foundation to develop this system.

The three-year project is led by Worcester Tech robotics engineering professors Cagdas Onal and Jie Fu seeking a robotic system that can navigate through messy, real-life environments filled with rough, uneven, and damaged surfaces filled with obstacles and clutter. The researchers envision a soft, segmented device that moves like a snake, with the intelligence to visualize, interpret, and adapt to its surroundings.

“The goal of this project is to have a snake-like robot with the autonomy to navigate through its environment without close supervision,” says Fu in a university statement. “In search-and-rescue scenarios, you’d want a robot you can rely on to go into areas inaccessible by humans and provide critical information about conditions there and about who needs to be rescued first.”

The device is expected to have a series of modules made of soft silicone rubber, each with its own motors, sensors, and control units that allow the system to move on its own. The modules would be connected into an integrated system, directed by two sets of algorithms. One group of algorithms will control the basic motions of the modules, responding to stimuli in the immediate environment.

In some cases, Onal explains, the device could use objects in its environment to move ahead. “A snake, for instance,” says Onal, “could move by wrapping part of its body around a rock so it can push against it and propel itself better. If it’s in a pipe or a maze-like structure, it might push off the walls to be able to move around an obstacle.”

The second set of algorithms will provide high-level planning and control, including coordinating the movements of individual modules in the device. The algorithms would make it possible for the device to learn from the accumulated experience of the device, translated into models for encountering future tasks.

“If I put the robot in the sand and then on gravel, it would exhibit different behaviors,” adds Fu. “On the higher level we can introduce learning-based mechanisms so the robot will learn the best way to move based on the environment it’s in and based on what it needs to achieve in that environment.”

While the Worcester Tech team expects its device to be used primarily in emergency situations like natural disasters, the researchers believe the system has lessons for industry, particularly in designing interactions between humans and robots. Unlike most robots, this device seeks out physical contact with its surroundings, which can result in robotics better designed for interactions with humans, and improve the ability of workers and robots to collaborate.

More from Science & Enterprise:

• Origami Robotics Designed for Fragile Objects
• University Spins Off Educational Robot Start-Up
• Exosuit Shown to Improve Walking for Stroke Patients
• Robot Designed to Grow into Shapes, Spaces
• Computer Vision, Deep Learning Aid Prosthetic Hands

*     *     *