While working at NASA in 2003, Dr. Robert Ambrose, director of the Robotics and Automation Design Lab (RAD Lab), designed a robot with no fixed top or bottom. A perfect sphere, the RoboBall could not flip over, and its shape promised access to places wheeled or legged machines could not reach — from the deepest lunar crater to the uneven sands of a beach. Two of his students built the first prototype, but then Ambrose shelved the idea to focus on drivable rovers for astronauts.
When Ambrose arrived at Texas A&M University in 2021, he saw a chance to reignite his idea. With funding from the Chancellor’s Research Initiative and Governor’s University Research Initiative, Ambrose brought RoboBall back to life.
Now, two decades after the original idea, RoboBall is rolling across Texas A&M University.
Driven by graduate students Rishi Jangale and Derek Pravecek, the RAD Lab is intent on sending RoboBall, a novel spherical robot, into uncharted terrain.
Jangale and Pravecek, both Ph.D. students in the J. Mike Walker ’66 Department of Mechanical Engineering, have played a significant part in getting the ball rolling once again.
Here is an exclusive Tech Briefs interview, edited for length and clarity, with Jangale and Pravecek.
Tech Briefs: What was the biggest technical challenge you faced while developing RoboBalls II and III? And did Dr. Ambrose give you any advice? How involved was he in the process?
Jangale: One of the biggest novelties of this robot is the soft shell. There exists a lot of spherical, pendulum-driven robots, but almost all of them have hard shells. So manufacturing a shell that was reliable and durable enough and then still approximates a sphere pretty closely, while the robot had any control authority, was really difficult.
Pravecek: Mechanically, it's very simple because there are only two joints. There are two things you have to deal with, where with traditional rovers, you can have three or four joints per leg and you can have six legs. It's a lot of motors, it's a lot of motor controller, it's a lot going on.
So mechanically and even electrically, it's pretty simple. You get that simplicity from the mechanics but then you apply that to the dynamics and how it behaves and how it moves and it gets really odd. Because the pendulum that's inside the vehicle is so heavy, it would be like you had a spaceship or a plane where the pilot weighed twice as much as the vehicle. And it becomes: How do you deal with the things inside of the thing you actually care about?
Then another technical challenge is troubleshooting the system — you can't see the robot because it's in the ball.
You really have to trust the telemetry you're putting across your network. You're looking at the data and you don't know how the pendulum or the insides are oriented. All you really have is what the sensors are telling you. It's not like a robot arm or like a rover where you can look at it and say, ‘I understand the problem.’ You just have to trust that your signal's right. And if something goes wrong, you have to take it all apart and look visually at what's happening.
Jangale: And I think another really important thing is that it's not really possible to serialize this development; we can't build the pendulum first, tune all our controls, and then go build the shell. They have to be done in parallel.
Dr. Ambrose helped a lot — giving us ideas, being able to bounce thoughts that we had off him. Just having another input to be able to gauge what might pan out, what won't.
Tech Briefs: Do you have any set plans for further tests, research, or work on the horizon?
Pravecek: Autonomy is a really interesting avenue for this kind of robot because we've made it such that the internal mechanism — the actuators, the sensors, the power — it's all isolated from the outside environment. It can have its own environment, kind of like a terrarium.
But with that, you cut off the outside and now the question is: How do I perceive the outside? So, we've added other modules to help with that, like GPS and RTK, and we’ve refined the algorithm that lets the robot know where it is. And then we're working on adding stereo cameras and different type of visual perception systems and LiDARs.
Jangale: One of the big advantages is that there's no wrong direction. You can roll in all the directions, and that gives you mobility over wheeled vehicles. But if you decide to put a payload on there that breaks the sphericality, now you’ve taken away one of the greatest advantages the robot has.
Tech Briefs: Those are all the questions I have. Is there anything else you'd like to add that I didn't touch up upon?
Jangale: It's been really fun to be able to take this robot and not be very gentle with it. A lot of people will build these robots, spend a lot of time and money, and then they're very careful, very systematic with their tests. They’re afraid to push it to its limits and as a result you don't necessarily get the full extent of understanding of what your robot can do.
But we don't really have those hesitations. We built this thing, we know what it does well, let's find out what happens if we push it off hills, go to the ocean.
Pravecek: It really speaks to the kind of environment Dr. Ambrose has fostered for us; he is incredibly comfortable with us trying things. Pushing the robots until they break and just learning as fast as we can.
Transcript
00:00:15 The Roboball started in 2003. We had kind of a nutty idea of this ball that could roll around, which we now call RoboBall I. And the the vision for that was a mission to the moon. And I'd seen a bunch of pictures of these craters on the moon that were really interesting that had had potentially materials on them, like water that would be real useful. But they were so big and so cold that I just knew we would never send a human down in there. So the idea was we'd roll a ball. I see a lot of applications, a number of roles that a ball can play. I think the opportunity on the moon is very real.
00:00:58 I'm really lucky to have such wonderful grad students here, and across all the projects, the A&M grad students are awesome. RISHI: A lot of the work that I do on this project is more on the hardware side, so I work kind of in the role of like a design engineer, where I work on, you know, maintaining the robot, designing its structure to make sure its performance is optimal and then, you know, performing upgrades as we learn things about how the robot wants to behave, how it performs, stuff at that. Goals that I have for Roboball are to see it be robust enough to tackle those missions without without having a lot of operator input and
00:01:34 being able to meet the performance criteria that a robot requires to be able to tackle that kind of terrain. I've always just been obsessed with the idea of seeing these these balls just, like, zooming. But when I first started RoboBall 3, I was super fast, but now, like, through a lot of time and testing and simulation and trying different things, we've got it up to 20 miles per hour. and we're trying to always go faster. So that's always my goal is to get these things to be move faster, smoother, more controlled. RISHI: This team is a really inspiring team to be on. It's by far, is some of the smartest people that I've ever met, and rather than everyone kind of working on their own goal, their own thesis,
00:02:13 it's a team of people who are solely interested in watching the ball get better and watching it progress. And so seeing everything everyone work so hard towards this common goal and put in their time, their effort, their ideas and support everyone else on the team, and their contributions to the robot makes it really rewarding and inspiring to be on this team. DEREK: They really are great, and they really are smart, and know what they're doing. And it's just it's always fun, you know, building robots, controlling robots, testing, it's really rewarding to do it with your friends. In long term, I'm hoping by the time the space is to is finished building, and it has those moonscapes and the
00:02:49 Marscape, we can take the robots out there and simulate like a full mission and kind of just, you know, as much as you can without going to the moon, prove that these things do what we wanted them to do. DR. AMBROSE: You know, it's all about the students. And so I'm here to help those students, grow and learn, and along the way develop some new ideas that they take out of the world with them.
