MTi helps in predicting a thrown ball’s trajectory

Camera view with ball trajectory predictions and artificial horizon from Xsens MTi measurements Camera view with ball trajectory predictions and artificial horizon from Xsens MTi measurements
MTi helps in predicting a thrown ball’s trajectory MTi helps in predicting a thrown ball’s trajectory

The DLR is Germany’s national research center for aeronautics and space. The DLR’s Institute of Robotics and Mechatronics is the high-tech knowledge center for the development of intelligently controlled mechanisms (in particular robots, small satellites, airplanes, vehicles). The newest development with their mobile humanoid Rollin’ Justin is that it has the ability to catch balls.

Catching a ball with a hand is not easy, neither for humans nor for robots. A tight interplay of fast perception, a good catching strategy, body control and dexterity is needed. Hence, ball catching is an excellent test bed for a number of robotic key technologies.

The robot’s head is equipped with stereo cameras to precisely track the trajectory of the ball in 3D space. Berthold Bäuml said that the addition of an inner ear — or inertial measurement unit (IMU) — allows the robot’s head to follow the ball to the hand.

‘It’s the same thing all humans and animals have in their heads, which allows you to measure the orientation of the head — so when you move your head you still know your relative position to the environment because you track the movement of the head with this inner ear system,’ Bäuml said in the The Engineer Magazine.

During the flight the predictions continuously improve and a real-time path planner decides where, when, and in which configuration to catch the ball in a kinematically optimal way. For this an optimization problem (including simple collision avoidance) is repeatedly solved on an external computing cluster (32 cores).

The visual tracking is a cooperation with the group of Udo Frese at DFKI Safe and Secure Cognitive Systems. He said: “Vision and inertial sensing are tightly coupled in human perception. The Xsens MTi allows us to give robots a comprehensive sense of motion, too”. Berthold Bäuml, the project head at DLR, adds: “The Xsens MTi turned out to be a valuable tool for a wide range of robotic systems in our institute, ranging from crawling robots, walking machines up to the mobile humanoid Rollin’ Justin, due to its compact size, high precision, robustness and ease of use.”

Camera view with ball trajectory predictions and artificial horizon from Xsens MTi measurements Camera view with ball trajectory predictions and artificial horizon from Xsens MTi measurements
MTi helps in predicting a thrown ball’s trajectory MTi helps in predicting a thrown ball’s trajectory

X

Xsens newsletter

Signup today to stay updated on all major Xsens news and developments. Subscribe now