Sensors
Multi-Time-of-Flight
The electronics developer and time-of-flight specialist Becom Systems has presented an embedded multi-ToF solution. The multi-ToF platform for several ToF sensors working in parallel with a central computing unit is based on the deep learning-capable Nvidia Jetson TX2 module with Nvidia Pascal GPU as well as dual-core Nvidia Denver 2 and quad-core ARM Cortex-A5 CPU.
The platform makes it possible to integrate up to four different sensors and combine them in different constellations - such as 2D and 3D ToF sensors. In principle, any imaging sensor technology can be integrated. The platform therefore offers integrators, software providers and product developers a compact and flexible solution that is typically suitable for applications in human-machine interaction at terminals, in collaborative robot interaction (cobots), in the automotive sector and in obstacle detection for autonomous systems.
The use of four independent sensors is particularly important when a camera is obscured by an object or a person. In the field of mobile robotics, there are requirements for a time-of-flight system that can only be met to a limited extent with individual industrial cameras. On the one hand, the desire to keep everything in view - which ideally means aperture angles of 360 degrees - as well as the combination with other information, such as a sensor fusion of 3D data with color information or infrared images. These requirements can be met with the new Multi-ToF platform. Areas of application include obstacle detection, sensor fusion, people detection, driver monitoring and gesture control. pb









