Industrial operations have sometimes to be carried out in demanding environments, for example in forests, mines or quarries, where maps with sufficient accuracy do not exist and the GPS signal cannot be relied upon. In order to be able to operate in these challenging environments, SLAM technologies are the solution. With Unikie’s SLAM techniques vehicles can simultaneously locate themselves relative to some starting point and relative to surrounding objects. And at the same time, a map of the environment is made and your own Itinerary is drawn on that high accuracy self-made map.
SLAM will always use several different types of sensors, and the powers and limits of various sensor types have been a major driver of new algorithms. SLAM can utilize data from Lidar, Radar, IMU and car data, like steering wheel angle. SLAM requires successful sensor fusion of various data sources in order to produce robust and accurate conclusions.
SLAM is a function of Unikie AI Vision – machine vision and machine learning framework.
Unikie’s machine intelligence technologies are based on Unikie AI Vision product. AI Vision is a sophisticated, hardware independent Machine Vision and Machine learning framework, that enables fast development of various Real-Time applications. It contains a highly sophisticated library of AI enabled algorithms to implement features like object detection, tracking, classification depending on the application.
Unikie AI Vision includes a sensor fusion solution for real time detection. AI Vision is capable of continuous and real-time 3D modeling of the environment with centimeter-level accuracy, utilising – in addition to LiDAR – several other sensors such as stereo cameras, ultrasound radars, GPS, and acceleration transducers. An additional feature of AI Vision is the shape recognition with which moving persons, objects and other environmental obstacles can be identified and tracked also in extreme conditions.
Northern America and Middle East