The rc_visard sensor family enables robots to generate and process time and location-related data in real time. The sensors support a variety of robot applications, ranging from bin-picking to navigation. The four versions of the rc_visard feature two different baselines (65 mm, 160 mm) and a color or monochrome acquisition capacity.
The rc_visard 160 monochrome is optimized for applications with a viewing distance greater than 50 cm that do not require color differentiation. The rc_visard 160 monochrome relies on cameras that are three times more light-sensitive than color cameras; hence it reliably works even in poor/variable lighting conditions.
Using ego-motion estimations (VINS), the rc_visards determine their position and orientation with millimetric precision and very low latencies. The passive stereo sensor works in natural and artificial light as well as low-light conditions. Precise ego-motion data is generated reliably, even in case of vibrations.
All rc_visards come with the same on-board software package that can be further enhanced by optional components from the rc_reason software suite – e.g. SLAM, TagDetect or ItemPick. An intuitive web interface enables an easy set-up and configuration. Last, but not least, multiple sensors can easily operate without interference in the same work space.
Downloads and Links
Tool for the discovery of Roboception’s rc_visard sensors via GigE Vision. This tool is required for detecting the rc_visards and perform factory resets.
Source Code: https://github.com/roboception/rcdiscover
The ROS driver for the rc_visard provides rectified images, disparity, confidence and error images in ROS format and can convert disparity images on-the-fly into depth images and colored point clouds. All image-related parameters can be controlled via dynamic reconfigure parameters. Additionally, poses from the rc_visard’s dynamics interface can be published on TF and as pose messages with additional information.
Source Code: https://github.com/roboception/rc_visard_ros
Roboception convenience layer around GenICam and GigE Vision as well as the dynamics interface of the rc_visard. The packages contain C++11 libraries and headers as programming interface as well as command line tools and examples for accessing and controlling the rc_visard.