In the field of sensor-assisted handling technology, we summarize all those applications in which objects are identified and located with the help of a sensor or sensor information (e.g. 3D image processing sensor, force-torque sensor) or the robot movement must be monitored and controlled. The applications are mainly in the field of 3D-object recognition (bin picking) but also in sensitive joining or grinding.

3D-object recognition- versatile applications

Our 3D object recognition solutions enable precise part identification and detection, gripping point determination and collision-free path planning. The automation solutions for the "bin picking" can be used in a variety of ways to optimize your part supply or feed.

The key component of our systems is the in-house developed object recognition software ART-3D. Based on the sensor data, it determines not only

• the exact position and orientation of objects,
• but also the optimum robot path for removing the workpiece.

The basis for this is, for example, the geometry model of object, the gripper and the container. All ART-3D object recognition algorithms work with 3D sensor data.

The advantage compared to 2D algorithms:

The object to be removed, foreign objects and overlaps can be determined more precisely. This is because the additional dimension - the third dimension - takes all existing information into account. No information is lost.

The software prevents collisions

The integrated path planning module (with inverse kinematics) takes grippers, containers or other interfering contours into account. This gives you the assurance that the robot will not collide with the environment during part removal and cause damage to the system.

An additional plus: The object recognition software can be used on conventional PCs under Windows.

Standardized interfaces enable communication with active image sensors (e.g. runtime measuring, projecting or light section sensors from different manufacturers), cell control and a higher-level host/control station system.