Wednesday, April 14, 2021
Home Technology Soft-bodied robots designed by AI algorithms

Soft-bodied robots designed by AI algorithms

Any point on soft-bodied robots can, in theory, deform in any way possible

There are some tasks for which traditional robots like the rigid and the metallic kinds cannot accomplish. On the other hand, soft-bodied robots may be able to interact with people more safely or slip into tight spaces with ease. But robots need to know the whereabouts of all their body parts to reliably complete their programmed duties. That’s a tall task for soft-bodied robots that can deform in a virtually infinite number of ways.

To help engineers design soft-bodied robots that collect more useful information about their surroundings, MIT researchers have developed an algorithm. Allowing the soft-bodied robots better interact with its environment and complete assigned tasks, the deep-learning algorithm suggests an optimized placement of sensors within the robot’s body. The advance is a step toward the automation of robot design. “The system not only learns a given task, but also how to best design the robot to solve that task,” says Alexander Amini. “Sensor placement is a very difficult problem to solve. So, having this solution is extremely exciting”.

The research will be presented during April’s IEEE International Conference on Soft Robotics and will be published in the journal IEEE Robotics and Automation Letters. Co-lead authors are Amini and Andrew Spielberg, both PhD students in MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). Other co-authors include MIT PhD student Lillian Chin, and professors Wojciech Matusik and Daniela Rus.

A long-running challenge in robotics is creating soft-bodied robots that complete real-world tasks. Their rigid counterparts have a built-in advantage: a limited range of motion. Rigid robots’ finite array of joints and limbs usually makes for manageable calculations by the algorithms that control mapping and motion planning. Soft-bodied robots are not so tractable.

Soft-bodied robots are flexible and pliant. Rather than a bowling ball, they generally feel more like a bouncy ball. “The main problem with soft robots is that they are infinitely dimensional”, says Spielberg. “Any point on soft-bodied robots can, in theory, deform in any way possible”. That makes it tough to design soft-bodied robots that can map the location of its body parts. Past efforts have used an external camera to chart the robot’s position and feed that information back into the robot’s control program. But the researchers wanted to create a soft robot untethered from external aid.

“You can’t put an infinite number of sensors on the robot itself,” says Spielberg. “So, the question is: How many sensors do you have, and where do you put those sensors in order to get the most bangs for your buck?” The team turned to deep learning for an answer.

Novel neural network architecture was developed by researchers. The architecture optimizes sensor placement and learns to efficiently complete tasks. First, the researchers divided the robot’s body into regions called “particles.” Each particle’s rate of strain was provided as an input to the neural network. Through a process of trial and error, the network “learns” the most efficient sequence of movements to complete tasks, like gripping objects of different sizes. At the same time, the network keeps track of which particles are used most often, and it culls the lesser-used particles from the set of inputs for the networks’ subsequent trials.

To ensure efficient performance, the network, by optimizing the most important particles, also suggests where sensors should be placed on soft-bodied robots. For example: the algorithm, in a simulated robot with a grasping hand, might suggest that sensors be concentrated in and around the fingers, where precisely controlled interactions with the environment are vital to the robot’s ability to manipulate objects. While that may seem obvious, it turns out the algorithm vastly outperformed humans’ intuition on where to site the sensors.

Against a series of expert predictions, the researchers pitted their algorithm. The team asked roboticists, for layouts of three different soft-bodied robots, to manually select where sensors should be placed to enable the efficient completion of tasks like grasping various objects. Then to compare the human-sensorized robots to the algorithm-sensorized robots, they ran simulations. And the results weren’t close. “Our model vastly outperformed humans for each task, even though I looked at some of the robot bodies and felt very confident on where the sensors should go,” says Amini. “It turns out there are a lot more subtleties in this problem than we initially expected”.

Spielberg says their work could help to automate the process of robot design. In addition to developing algorithms to control a robot’s movements, “we also need to think about how we’re going to sensorize these robots, and how that will interplay with other components of that system,” he says. And better sensor placement could have industrial applications, especially where robots are used for fine tasks like gripping. “That’s something where you need a very robust, well-optimized sense of touch,” says Spielberg. “So, there’s potential for immediate impact”

“Automating the design of sensorized soft-bodied robots is an important step toward rapidly creating intelligent tools that help people with physical tasks,” says Rus. “The sensors are an important aspect of the process, as they enable the soft robot to “see” and understand the world and its relationship with the world”.

Also read: Fortinet Announces AI-powered XDR for Fully Automated Threat Detection, Investigation, and Response

Do Follow: CIO News LinkedIn Account | CIO News Facebook | CIO News Youtube | CIO News Twitter

khushbu
Chief Editorhttps://www.cionews.co.in
Chief Editor - CIO News | Founder & CEO - Mercadeo

LEAVE A REPLY

Please enter your comment!
Please enter your name here

- Advertisment -WhatsApp Image 2020 11 10 at 1.52.58 PMWhatsApp Image 2020 11 10 at 1.52.58 PMWhatsApp Image 2020 11 10 at 1.52.58 PMWhatsApp Image 2020 11 10 at 1.52.58 PM

Most Popular

Next-Generation Incubation Scheme: 42 start-ups eligible for benefits

Next-Generation Incubation Scheme: 42 start-ups eligible for benefits

Ed-Tech venture to offer end-to-end learning ecosystem

Ed-Tech venture to offer end-to-end learning ecosystem

Ransom-ware: A malware type used to extort money

Ransom-ware: A malware type used to extort money

Digital Ethics framework to be considered for responsible digitalisation

Digital Ethics framework to be considered for responsible digitalisation

Recent Comments