The July 2022 issue of IEEE Spectrum is here!

Close bar

The Secret to Small Drone Obstacle Avoidance Is to Just Crash Into Stuff

Small drones bumble through obstacles just like bees

3 min read
GRASP Lab pico drone can fly through obstacles
GRASP Lab's pico drones can bumble through obstacles just like bees.
Image: GRASP Lab/University of Pennsylvania

Roboticists are putting a tremendous amount of time and effort into finding the right combination of sensors and algorithms that will keep their drones from smashinginto things. It’s a very difficult problem: With a few exceptions, you’ve got small platforms that move fast and don’t have the payload capability for the kind of sensors or computers that you really need to do real-time avoidance of things like trees or power lines. And without obstacle avoidance, how will we ever have drones that can deliver new athletic socks to our doorstep in 30 minutes or less?

At the University of Pennsylvania’s GRASP Lab, where they’ve been working very, very hard at getting quadrotors to fly through windows without running into them, Yash Mulgaonkar, Luis Guerrero-Bonilla, Anurag Makineni, and Professor Vijay Kumar have come up with what seems to be a much simpler solution for navigation and obstacle avoidance with swarms of small aerial robots: Give them a roll cage, and just let them run into whatever is in their way. Seriously, it’ll be fine!

GRASP Lab pico dronesImage: GRASP Lab/University of Pennsylvania

This kind of “It’ll be fine” philosophy is what you find in most small flying insects, like bees: They don’t worry all that much about bumbling into stuff, or each other; they just kind of shrug it off and keep on going. Or, if you’re a roboticist, you might say something like, “The penalty due to collisions is small at these scales, and sensors and controllers are not precise enough to guarantee collision-free trajectories,” so stop trying to solve the collision problem, and just focus on not completely trashing yourself when you hit something. (Swiss startup Flyability was among the first to demonstrate the benefits of collision robustness by equipping a regular-size drone with a gimballed protective cage and flying through forests and ice caves.)

In designing its flying vehicles, the UPenn group sought a bio-inspired approach, focusing on small and resilient quadrotors. The group built a fleet of 25-gram, 10-centimeter-wide pico quads, each featuring a lightweight, Gömböc-inspired self-righting roll cage made from a heat-cured yarn consisting of 12,000 strands of carbon fiber. The pico quads are piloted by a simple controller that “does not consider the position of other robots or obstacles, and does not contain special actions to command as a consequence of a collision,” the researchers wrote in a recent paper. All it does is try to stabilize the pico quads as best it can, while directing them toward a goal position. This works impressively well, even when the robot has zero knowledge of any obstacles or other pico quads that might be in its way:

“This is a radically different approach to flight where you simply rely on local information to navigate and build in the resilience to collisions that are inevitable because of noisy sensors and resource-constrained processors,” Vijay Kumar tells us. “We are trying to make robots that are safe and smart and that can operate in cluttered indoor environments.” In terms of applications where this kind of capability would be most useful, Kumar mentions as one example exploring an indoor environment during an emergency-response situation:

Imagine a search-and-rescue or disaster response scenario when you want a swarm of robots to enter a contaminated or dangerous building to create maps. The robots that [we have] created allow you to explore simple control algorithms that no longer require clever approaches to solving the exponentially hard problem of motion planning. If you allow robots to collide, you can use very simple algorithms for navigation.

“Bio-inspired Swarms of Small Aerial Robots,” by Yash Mulgaonkar, Luis Guerrero-Bonilla, Anurag Makineni, and Vijay Kumar from the University of Pennsylvania, has been submitted to Interface Focus.

[ Vijay Kumar Lab ]

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

“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