Monthly Newsletter | 1st December 2016

IN THIS ISSUE

  • Current Article
  • News
  • Quote
  • Previous Issue
  • Facebook Posts
  • Contact us
Write to Us at contactus@aimumbai.org

MAKING A ROBOT WALK

Goes a Bulgarian saying, "From Walking -- Something, From Sitting -- Nothing". Walking is perhaps one of the most important traits, in anyone's life. It distinguishes and at times decides the ability of an individual. Humans, born with a learning and sharp brain, eventually adapt to the surroundings and slowly and steadily learn to walk. The brain, being so advanced, handles all of the mechanical calculations associated with walking, without us having to consciously do it. But what in case of Robots....?

And, what good is a robot, if it can't ambulate in any way. How do you teach a robot to walk? In order to do that we need to be 'consciously' introduced to the procedure of 'walking'.

“In walking the heel is first placed on the ground; the weight is next thrown on the ball of the foot, and the body is raised so as to permit the free limb to swing by the one upon which the body rests. As soon as the free limb has passed the center of gravity, the body is allowed to descend, until the heel on that side receives the weight, when the body is again raised.”
- Scientific American

In case of a robot, every component of the robot contributes to the CG of the robot.(CG or Center of Gravity is the point at which the weight is evenly distributed and all sides are in balance.) This CG is responsible for whatever it is, that balances itself or stands. It gives stability to any structure. In case of walking, due to the constant movement, the CG keeps shifting at any given point. Therefore, having a well balanced body posture, which doesn't hamper the CG while the robot is walking or standing is of paramount importance.

Another feature which affects walking is 'friction'. Friction in simple terms is the resistance offered by the surface, which in our case, will help the robot to keep its foot stable while the other one is being put forward. In other words, the surface on which the robots walks, needs to have just enough friction, to restrict the foot from any movement while the robot's weight is balanced on it and the other leg is being swung across in the front. Therefore, a proper combination of CG and friction are two things to be taken care of, in case of making a robot walk.

Doing this would just make the robot ready to walk in a controlled environment, which even the human brain requires to take care of. But a robot would be really intelligent if it is able to handle any external forces and still be in the initial position. The most widely adopted way to do this is by using ZMP(Zero Moment Point). In this approach, robots rely on force sensors installed on their feet, and a feedback control loop that constantly adjusts their position relative to this stable point (the ZMP) to maintain balance. A position-based compliance control that allows the robot to adapt to external disturbances requires force measurement at every expected contact point, which increases the computational load and creates unavoidable time delays.

A more recent study by Christian Ott and his team, related to the same problem statement tries to address it, with a more effective and reasonable approach. It uses a formulation coming from the field of robot grasping, which considers at the same time a desired force and torque that allow the robot to recover the initial position and orientation [when disturbed]. This, grasping-inspired controller elegantly takes into account friction at the contact points with the ground. First, it uses an optimization algorithm that computes the forces needed at each point to neutralize the perturbation. Then it determines how to move the robot's torque-controlled joints to produce the desired forces. The result: equilibrium. The challenge, here, is to use the same methodology while walking. Also, this method currently uses a finite set of predetermined contact points, but the researchers are confident that they can extend it to a more general model.

It took centuries for monkeys to evolve and attain the distinguishable feature of walking on two feet and evolve into human. But the evolution of robots has been a journey of a few decades. So can robots be the next superspecies to take over the world and create a more evolved living than we have on two feet....

News

Google Translate’s AI can translate between languages where it wasn’t taught to do so
Google's Translate's AI has created its own secret language to translate between languages where it wasn't taught to do so.
New AI therapy to help overcome fear: study
Scientists have developed a method to read and identify a fear memory using a new technique called 'Decoded Neurofeedback'.
Tesla’s More Advanced Autopilot Arrives Next Month
Autonomous driving on highways is nothing. Take a look at Tesla's new more advanced hardware and software and what it can do in a busy suburban environment.
Quote

““If an AI possessed any one of these skills—social abilities, technological development, economic ability—at a superhuman level, it is quite likely that it would quickly come to dominate our world in one way or another. And as we’ve seen, if it ever developed these abilities to the human level, then it would likely soon develop them to a superhuman level. So we can assume that if even one of these skills gets programmed into a computer, then our world will come to be dominated by AIs or AI-empowered humans.””

― Stuart Armstrong (Author of Smarter Than Us: The Rise of Machine Intelligence)

Previous Issues

Decoding Artificial Intelligence, Machine Learning, Deep Learning

Dated: 1 November 2016

Genetic Algorithms: An Introduction With Example

Dated: 3 October 2016

View All
While you were away
The first DJ powered by Artificial Intelligence
From playing your morning wake up the tune to mood swinging music, and from creating your perfect party playlist to hands-free voice commands. #Hurricane, the first DJ powered by #AI is going to be one of the best mp3 players.
Isn't it?
Autonomous Cars and Human Drivers Will Battle for Ride-Sharing Jobs
As we earlier shared our thought on a battle over the heat blowing topic, Does Artificial Intelligence is going to eat jobs or is going to provide more job opportunities.? This one is from the inventor, CEO as well as CTO of SpaceX, Elon Musk.
First International Beauty Contest Judged by Artificial Intelligence Beauty.AI 1.0
All of those who work hard to be beautiful and to be the beauty contest winner, be alert.
Because your beauty contest is going to get judged by an #AI. Are you ready?
Contact Us

Your reviews are valuable. Help us to improve.
Give your feedback on contactus@aimumbai.org
www.aimumbai.org

Address: 10, Jamanadas Industrial Estate, Phase -1, Dr. R.P. Road, Mulund (West),, Mumbai, Maharashtra 400080


© Copyright Artificial Intelligence Mumbai. All Rights Reserved.