The global site of the UK's leading magazine for automation, motion engineering and power transmission
13 July, 2024

Twitter link

Robot can catch a tumbling bottle in mid-flight

14 May, 2014

Swiss researchers have programmed a robot to catch a variety of spinning objects – including tennis rackets, hammers and half-full bottles – when they are thrown towards it.

The robot, developed by researchers at the École polytechnique fédérale de Lausanne (EPFL), can grasp objects with complex shapes and trajectories in less than five hundredths of a second.

The 1.5m-long Kuka robot arm has three joints and a sophisticated hand with four fingers. The application, programmed at EPFL’s Learning Algorithms and Systems Laboratory (Lasa), requires several parameters to be integrated and for the arm to react rapidly to unforeseen events.

Existing robotic systems have limitations for tasks of this kind. “Today’s machines are often pre-programmed and cannot quickly assimilate data changes,” explains Lasa’s head, Aude Billard. “Consequently, their only choice is to recalculate the trajectories, which requires too much time from them in situations in which every fraction of a second can be decisive.”

She points out that robots will increasingly be needed to catch or dodge complex objects in full-motion. For example, they could be used to protect humans by catching falling objects in dangerous locations. “Not only do we need machines able to react on the spot, but also to predict the moving object’s dynamics and generate a movement in the opposite direction,” she says.

The Lasa researchers were inspired by the way humans learn: by imitation, and trial and error. This approach, called “programming by demonstration”, does not give specific instructions to the robot. Instead, it is shown examples of possible trajectories. The arm is guided manually to the projected target and this exercise is repeated several times.

The research was conducted with objects including a ball, an empty bottle, a half-full bottle, a hammer and a tennis racket. These objects were chosen because the part of the object that the robot has to catch – the handle of the tennis racket, for example – does not correspond to its centre of gravity. When thrown into the air, these objects make complex movements, often involving several axes.

Catching a half-full bottle poses an extra challenge because its centre of gravity changes during its trajectory.

During an initial learning phase, the objects are thrown repeatedly towards the robot. Using an array of cameras mounted around it, the robot creates a model for the objects’ kinetics, based on their trajectories, speeds and rotational movement.

This model is then translated into an equation that allows the robot to position itself rapidly in the right position whenever an object is thrown towards it. During the first few milliseconds of flight, the machine refines and corrects the trajectory for a precise, real-time capture. This efficiency is enhanced further by using controllers that couple and synchronise the movements of the robot’s hand and fingers.

  • To view a digital copy of the latest issue of Drives & Controls, click here.

    To visit the digital library of past issues, click here

    To subscribe to the magazine, click here



"Do you think that robots create or destroy jobs?"



Most Read Articles