The July 2022 issue of IEEE Spectrum is here!

Close bar

Shrewbot Uses Whiskers to Map Its Environment

A robot modeled after a shrew can figure out where it is using just its whiskers

2 min read
Shrewbot Uses Whiskers to Map Its Environment

Robots that make maps tend to be highly reliant on vision of one sort or another, whether it’s a camera image or something off the end of the visible spectrum like a laser scanner. This is understandable: humans are adapted to use vision, so we understand it pretty well, and we can get a lot of useful information out of a visual image. Animals, on the other hand, take advantage of a much broader suite of senses, specialized for their environments. If you only come out at night, or if you live in a hole, vision is perhaps not the best solution for you, and a robot modeled after a shrew can now make maps using just tactile feedback from a prodigious set of artificial whiskers.

We met Shrewbot in January of last year; it’s an adorable robot with artificial whiskers modeled after the Etruscan pygmy shrew. Here’s the video from 2012:

New research presented at the IEEE International Conference on Robotics and Automation (ICRA) this week has Shrewbot performing what the researchers are calling tSLAM, which is tactile Simultaneous Localization and Mapping. The robot has an array of 18 individually-actuated whiskers mounted on a 3 degree-of-freedom neck, attached to an omni-drive mobile platform. Using a combination of wheel odometry and detection by whisking (the behavior really is called whisking), Shrewbot is able to gradually make a tactile map of an area by combining hundreds (or thousands) of whisk contacts that it feels when it encounters walls or other obstacles.

This video shows the mapping in action; the blue line shows the actual location of the robot, while the red line shows where the robot thinks it is. Notice how the red line more closely matches the blue line as the robot makes more whisk contacts:

By the end of this process, you can see that Shrewbot has a reasonably good idea of what its environment looks like. And remember, the resulting map (and ability to localize) is achieved purely through touch. Robots like Shrewbot are ideal for exploring and mapping spaces where laser, acoustic, or visual sensors don’t work very well, like dark spaces, spaces filled with dust or smoke, or even in turbid water, and future research will investigate how well this technique works at larger scales, with an eye towards practical deployment, and perhaps even an implementation of texture detection with whiskers as well.

“Simultaneous Localization and Mapping on a Multi-Degree of Freedom Biomimetic Whiskered Robot,” by Martin J. Pearson, Charles Fox, J. Charles Sullivan, Tony J. Prescott, Tony Pipe, and Ben Mitchinson from the Bristol Robotics Laboratory and Sheffield University's Adaptive Behaviour Research Group, was presented this week at ICRA 2013 in Germany.

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

“I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

This article is part of our special report on AI, “The Great AI Reckoning.”

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓Show less