MIT’s Cheetah 3 is very light on its feet, it can now “leap and gallop across rough terrain, climb a staircase littered with debris, and quickly recover its balance when suddenly yanked or shoved,” all while it doesn’t have cameras to track its surroundings, reported MIT News.
Weighing in at 90 pounds, the cheetah robot is designed to navigate almost any kind of terrain without tracking camera. Instead, the robot “feels” its surroundings, as described by MIT engineers as “blind locomotion,” sort of like feeling for a light switch in the dark.
“There are many unexpected behaviors the robot should be able to handle without relying too much on vision,” says the robot’s designer, Sangbae Kim, associate professor of mechanical engineering at MIT. “Vision can be noisy, slightly inaccurate, and sometimes not available, and if you rely too much on vision, your robot has to be very accurate in position and eventually will be slow. So we want the robot to rely more on tactile information. That way, it can handle unexpected obstacles while moving fast.”
In the event a motor or limb malfunctions, the cheetah is designed with modular components: Three electric motors power each of the robot’s legs. Each motor can easily be swapped out for a new one, or even the leg can be replaced.
MIT engineers will present the cheetah’s vision-free capabilities in October at the International Conference on Intelligent Robots, in Madrid.
Kim expects in the future that the robot could conduct tasks that would otherwise be too dangerous or inaccessible for humans.
“Cheetah 3 is designed to do versatile tasks such as power plant inspection, which involves various terrain conditions including stairs, curbs, and obstacles on the ground,” Kim says.
“I think there are countless occasions where we [would] want to send robots to do simple tasks instead of humans. Dangerous, dirty, and difficult work can be done much more safely through remotely controlled robots.”
According to MIT News, the robot can blindly walk up staircases and through unstructured terrain and can immediately recover its balance if unexpected forces knocked it over. This capability is due to two new algorithms developed by Kim’s team: a contact detection algorithm, and a model-predictive control algorithm.
The contact detection algorithm supports the robot in decision making for the best time to jump or step around an object. For example, if the robot steps on a twig versus a large rock, it will understand how to react and make the proper adjusts so that it can continue forward.
“When it comes to switching from the air to the ground, the switching has to be very well-done,” Kim says. “This algorithm is really about, ‘When is a safe time to commit my footstep?'”
The algorithm calculates these probabilities based on data from gyroscopes, accelerometers, and joint positions of the legs, which record the leg’s angle and height concerning the ground.
Engineers tested the algorithm in experiments with the robot trotting on a laboratory treadmill and climbing on a staircase. Both surfaces had irregular objects placed on it to simulate a construction site.
“It doesn’t know the height of each step, and doesn’t know there are obstacles on the stairs, but it just plows through without losing its balance,” Kim says. “Without that algorithm, the robot was very unstable and fell easily.”
The model-predictive control algorithm calculates the robot’s body and legs a half-second into the future.
In tests, researchers introduced unexpected forces by physically abusing the robot with kicks and shoves as it trotted on a treadmill, and even pulled on a leash to make it fall down a staircase.
Engineers discovered that the model-predictive algorithm allowed the robot to instantly create counter-forces to regain the center of balance and keep moving forward after a fall.
Engineers have also added tracking cameras to some of the robots to give it an understanding of its surroundings. The cameras will enable it to map out its environment and pre-determined plan of action for more considerable obstacles such as doors and walls. Engineers told MIT News that they’re further progressing the robot’s blind locomotion capabilities.