World leader in AI software for vision guided robotics
We guide you through an easy and robust 3-step process.
1. Data acquisition
A suitable 3D-camera is mounted to take images of the items in the actual picking environment.
2. Training of software
Complementing our existing data set, the new images are used to optimize the training of our neural network for the client’s specific application. The performance and reliability of the system software are also evaluated.
A locally installed PC takes images as input between each cycle of the robot. The optimized algorithm instantly proposes the best possible 6DoF grasp poses, as well as additional relevant information about the type of object.