RoboSense (Suiteng Juchuang) vehicle gauge-level solid-state lidar L4 perception fusion scheme released: perception radius up to 200m

Thank you, Mr. Air, a netizen of CTOnews.com, for your clue delivery! CTOnews.com March 3 news, RoboSense recently released the first vehicle-class solid-state lidar L4 perceptual fusion scheme RS-Fusion-P6 (P6). The P6 is equipped with M-series vehicle gauge-level solid-state lidar, self-developed point cloud sensing software Perception, and integrated hardware such as a variety of sensors and computing units.

The scheme adopts 4 vehicle gauge-level solid-state lidar fusion, 7 types of dynamic and static obstacle sensing coverage (pedestrians, carts, trolleys, trailers, cyclists, conical bucket-stone pier lanes, kerbs, etc.), 360X25 °field of view angle FoV,2 road marking sensing coverage (lanes, kerbs, etc.), with a maximum perception radius of 200m (detection of driveable areas based on global obstacles and road signs distribution) 52 obstacle perception data output (ID, type, motion state, size, position, speed, etc.).

Support LiDAR one-stop flexible deployment, integrate four RS-LiDAR-M series lidars for the first time to provide more excellent field of view and sensing distance (CTOnews.com Note: can flexibly choose 8 M series lidars to be deployed anywhere on the body), the horizontal range of intelligent sensing 360 °can see wider, the farthest sensing radius is 200m, and the distance is even farther.

Using point cloud algorithm technology, multi-lidar perception results are output independently, and real-time fusion processing is carried out to accurately restore complex traffic environment information. In terms of software, P6 carries the real-time point cloud perception software Perception developed by Suiteng Juchuang, which can provide real-time point cloud perception results and has rich and perfect perception functions. Whether it is 4 or more M-series lidars, Perception can process raw perceptual data with high quality and efficiency.

RS-Fusion-P6 can flexibly respond to a variety of traffic driving scenarios, including mixed pedestrian and vehicle scenarios, accurately distinguish all kinds of traffic participants and wait for safe traffic opportunities; unprotected left-turn scenarios, perceive lateral coming cars in advance, turn left more decisively and safely; import traffic scenarios, accurately sense the position and speed of vehicles behind the side, and quickly merge into traffic flow; vehicle congestion scenarios, accurately predict the emergency congestion of vehicles in front to avoid collisions.