Skip to content
Last Dictionary

locomotion

nounAlso: robot locomotion, legged locomotion, loco-manipulation

In plain wordsHow a robot moves itself from place to place: rolling on wheels, walking on legs, crawling, swimming or flying.

  1. The ability of a robot or animal to move its whole body through an environment, and the engineering of that ability. Robots locomote with wheels and tracks, which are efficient on prepared ground; with legs, which can step over obstacles, climb stairs and cross rough terrain at the cost of balance and control complexity; and by crawling, swimming, flying or hybrids such as wheeled legs. Legged locomotion is the most studied in AI because it requires continuous balance. Classical controllers rely on simplified dynamics such as the zero moment pointzero moment point, n.: The spot on the ground beneath a walking robot where the floor's push balances out; keeping it inside the area under the robot's feet stops the robot from tipping over. criterion and inverted-pendulum models, combined with model predictive controlmodel predictive control, n.: A control method that repeatedly plans the best moves a few seconds ahead using a model of the machine, carries out only the first move, then plans again with fresh measurements.; since about 2019, controllers trained by reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time. in a physics simulatorphysics simulator, n.: Software that imitates how real objects move, collide, bend and fall, so robots and their AI can be designed, trained and tested on a computer before touching the real world. with domain randomizationdomain randomization, n.: Training a robot's AI in a simulation whose colors, lighting, weights, friction and other details are shuffled at random on every run, so that the real world holds few surprises it has not already practiced for. and transferred to hardware (sim-to-real transfersim-to-real transfer, n.: Training a robot's AI in a computer simulation, where practice is fast, cheap and safe, and then making it work on the real robot despite the differences between simulation and reality.) have become standard for quadrupeds and humanoidshumanoid robot, n.: A robot built roughly in human shape, with a head, torso, two arms and usually two legs, so that it can work in spaces and with tools designed for people., producing robust walking on snow, rubble and stairs and parkour-style jumps. Whole-body ('loco-manipulation') control combines locomotion with manipulationmanipulation, n.: A robot's use of arms, hands or grippers to change the world around it: picking things up, moving, pushing, opening, assembling or using tools..

    Locomotion is the easy part now; the robot falls when it tries to open the door while walking.

Origin

From Latin loco, 'from a place', and motio, 'motion'. Robotic legged locomotion was pioneered by Ralph Mosher's GE Walking Truck (1960s), Waseda University's WABOT-1 (1972) and Marc Raibert's dynamically balancing hopping robots of the 1980s. Learned locomotion took off with sim-to-real work such as Jie Tan and colleagues' Minitaur gaits (2018) and Jemin Hwangbo, Marco Hutter and colleagues' controllers for the ANYmal quadruped (2019).

Related

See also
  • legged robotlegged robot, n.: A robot that walks on legs instead of rolling on wheels, such as a four-legged 'robot dog' or a two-legged humanoid, so it can climb stairs and cross rough ground.
  • humanoid robothumanoid robot, n.: A robot built roughly in human shape, with a head, torso, two arms and usually two legs, so that it can work in spaces and with tools designed for people.
  • mobile robotmobile robot, n.: A robot that travels around by itself — on wheels, tracks, legs, propellers or thrusters — such as a robot vacuum, warehouse cart, delivery robot or planetary rover.
  • zero moment pointzero moment point, n.: The spot on the ground beneath a walking robot where the floor's push balances out; keeping it inside the area under the robot's feet stops the robot from tipping over.
  • sim-to-real transfersim-to-real transfer, n.: Training a robot's AI in a computer simulation, where practice is fast, cheap and safe, and then making it work on the real robot despite the differences between simulation and reality.
  • model predictive controlmodel predictive control, n.: A control method that repeatedly plans the best moves a few seconds ahead using a model of the machine, carries out only the first move, then plans again with fresh measurements.
  • manipulationmanipulation, n.: A robot's use of arms, hands or grippers to change the world around it: picking things up, moving, pushing, opening, assembling or using tools.
Contrast with
  • manipulationmanipulation, n.: A robot's use of arms, hands or grippers to change the world around it: picking things up, moving, pushing, opening, assembling or using tools.

Sources

  1. Raibert, M. H. (1986). Legged Robots That Balance. MIT Press.
  2. Hwangbo, J. et al. (2019). Learning agile and dynamic motor skills for legged robots. Science Robotics 4(26).

Mentioned in

humanoid robothumanoid robot, n.: A robot built roughly in human shape, with a head, torso, two arms and usually two legs, so that it can work in spaces and with tools designed for people., robotrobot, n.: A machine that senses its surroundings, decides what to do and physically acts, moving itself or handling objects, with some independence from a human operator. Factory arms, robot vacuums and self-driving cars are all robots., roboticsrobotics, n.: The science and engineering of building robots and making them work: designing their bodies, sensors and motors, and writing the software that lets them see, plan and move safely., legged robotlegged robot, n.: A robot that walks on legs instead of rolling on wheels, such as a four-legged 'robot dog' or a two-legged humanoid, so it can climb stairs and cross rough ground., manipulationmanipulation, n.: A robot's use of arms, hands or grippers to change the world around it: picking things up, moving, pushing, opening, assembling or using tools., model predictive controlmodel predictive control, n.: A control method that repeatedly plans the best moves a few seconds ahead using a model of the machine, carries out only the first move, then plans again with fresh measurements., sim-to-real transfersim-to-real transfer, n.: Training a robot's AI in a computer simulation, where practice is fast, cheap and safe, and then making it work on the real robot despite the differences between simulation and reality., domain randomizationdomain randomization, n.: Training a robot's AI in a simulation whose colors, lighting, weights, friction and other details are shuffled at random on every run, so that the real world holds few surprises it has not already practiced for., zero moment pointzero moment point, n.: The spot on the ground beneath a walking robot where the floor's push balances out; keeping it inside the area under the robot's feet stops the robot from tipping over.