Keywords: LSLiDAR, LiDAR, Dust and Fog Recognition, Point Cloud Algorithm, CH128X1, C16/C32, Perception Accuracy, Data Stability, Indoor and Outdoor Applications, Industry Application
When discussing LiDAR, the first things that come to mind are detection distance, motion state, identification, resolution, and tracking of targets. In practical applications, adverse weather conditions such as dust, rain, and fog often interfere with the operation of LiDAR.
LiDAR has diverse demands in different application scenarios. Indoor environments, like warehouses where unmanned forklifts or robots are moving goods, could be high in dust. Outdoor environments require LiDAR to adapt to various conditions, particularly when vehicles are faced with dust and foggy weather. Under these conditions, the operation of lidar can be unstable, leading to diminished detection capabilities or the output of unnecessary interference point clouds, thereby challenging the reliability of LiDAR.
Indeed, LiDAR faces interference from dust and fog in real-world applications. To address these issues, LSLiDAR has upgraded its auto-grade hybrid solid-state LiDAR CH128X1 and multi-line mechanical LiDAR C16/32, equipping them with dust and fog identification and filtering capabilities.
CH128X1 Auto-grade 905nm Hybrid Solid-state LiDAR
LSLiDAR has conducted extensive comparative experiments with dust characteristics and point cloud algorithms, developing a LiDAR dust filtering algorithm suitable for mining truck operation conditions. This algorithm, in combination with any LSLiDAR’s LiDAR mounted on a vehicle, can be optimized and enhanced multiple times based on customer needs and actual application scenarios. The algorithm and data learning breakthrough blind spots, solving the problem of identifying targets and perception accuracy behind the dust filter, increasing operational efficiency, data stability, and safety. Unhindered by environmental dust and fog, it provides reliable point cloud information indoors and outdoors, assisting in accurate target identification, navigation, and obstacle avoidance, yielding groundbreaking results.
GIF 1 Dust Filter Features Off
GIF 2 Dust Filter Features On
GIF 3 Rainwater Filter Features Off
GIF 4 Rainwater Filter Features On
As can be seen from GIF images 1-2, there is a lot of dust around or in front of LiDAR. However, the actual point cloud data of LiDAR shows that it can penetrate thick dust and detect targets behind it, even when the dust point cloud is significantly reduced. In images 3-4, during rainfall, scattered raindrops are treated as obstacles, hindering the path of vehicles or robots. The radar removes these raindrop points, relieving the data processing burden of the host computer. As such, CH128X1 provides more accurate perception and higher sensitivity.
C16/32 Multi-line Mechanical LiDAR
LSLiDAR’s multi-line mechanical LiDAR C16/C32 has undergone a 4.0 upgrade, offering higher detection accuracy, more stable point clouds, and stronger light interference resistance. Additionally, it has been repeatedly tested for dust and fog, providing effective dust recognition and filtering, taking perception performance to a new level.
C16/C32 Effect Comparison
LSLiDAR full-scene LiDAR, with its high stability and reliability, is applied in various fields and has been recognized by many well-known customers in the industry. LSLiDAR is committed to solving customer pain points, continuously adjusting and optimizing to provide safe, efficient, and high-value perception products.
