The MIT researchers ring in the good news of steady progress in their robotics projects as they announce an improved design of its MIT Cheetah 3 robot to the world. So, the 90-pound Cheetah robot, albeit blind, can now brush past rough terrain just like in the natural world, climb staircases laden with debris, regain balance after being disturbed by an external force and so on. All this can be done by it without external environmental sensors or cameras, likening it to the nature and sharp instincts of the Labrador, which is what the mechanical Cheetah is like in size. This ability to feel its way through pitch-black darkness is touted by the engineers as “blind locomotion.
Sangbae Kim, associate professor of mechanical engineering, MIT, stated “There are many unexpected behaviors the robot should be able to handle without relying too much on vision,” says the robot’s designer. “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.”
Up ahead in October, the International Conference on Intelligent Robots in Madrid is slated to be the venue for unveiling the robot’s vision-free capabilities, advanced hardware, including a wider range of motion (backwards and forwards stretching and side-to-side twists) as compared to Cheetah 2, establishing a stepping stone for further improvements in the robot to perform tasks in future that are beyond human capacity.
Kim’s statement on this goes: “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. 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.”
The balance-recovering ability of the Cheetah 3 can be attributed to a contact detection algorithm, and a model-predictive control algorithm developed by Kim and his team, that enables the robot to ascertain the best time to place its swinging leg on the ground, ie switch positions intelligently in the nick of time as per the situation, which is crucial for the trying circumstances where the leopard is to do its tasks.
When it comes to switching from the air to the ground, the switching has to be very well-done .This algorithm is really about, ‘When is a safe time to commit my footstep? – Sangbae Kim, associate professor of mechanical engineering, MIT
The robotics enthusiasts can look up the post on MIT news to get exacting information on the Cheetah 3 project.
As for the robot’s ‘blind locomotion’, the model-predictive control algorithm is responsible for it partially, predicting the amount of force a leg should apply on its step. Moreover, this algorithm also calculates the multiplicative positions of the robot’s body and legs one half-second into the future, given that a certain force is applied on the ground by a leg.
For surveillance and mapping of the general environment, the team had installed cameras in the robot to keep the visual alerting mechanism alive, such as on doors and walls, but the priority of the team is to constantly work upon and improve the robot’s blind locomotion to this end, Kim said: “We want a very good controller without vision first. And when we do add vision, even if it might give you the wrong information, the leg should be able to handle (obstacles). Because what if it steps on something that a camera can’t see? What will it do? That’s where blind locomotion can help. We don’t want to trust our vision too much.”
With the design of the Cheetah robot improved one notch higher, it’s nothing short of commendable work by the MIT researchers’ team to produce a dynamic mechanical beast which can pounce into dangerous, disaster-struck and humanly inaccessible environments, winning over obstacles that can’t be dealt with by us, the humans.
Source: MIT News