For this collaboration, Livox completed a series of customizations for Xpeng using its Horiz sensor (automotive-grade version of Livox Horizon), with its final mass-produced version boasting a range of industry-leading features.
For example, the detection range of Horiz will be increased to 150m from the 90m of the public sample version Horizon (for objects at 10% reflectivity). This will enable Xpeng’s XPILOT autonomous driving system to detect any remote obstacle much earlier with absolute ease while on highways and urban roads.
Meanwhile, the resolution of Horiz also doubles its public sample version, with the new ROI (Region Of Interest) point cloud density increased to 144 lines equivalently at 0.1 second without the need for adding extra laser transmitters.
The increased point cloud density enables the faster detection of tiny objects on the road surface, including pedestrians, bicycles or even traffic cones. The horizontal FOV (Field Of View) of Horiz has also been expanded to 120° from the 81° of Horizon. This greatly enhances the smart driving experience by resolving many persistent challenges faced by drivers, including the removal of blind spots against cut-in vehicles.
Comparison of FOV and scanning pattern between Horizon and Horiz
Livox’s customized solution for Xpeng also includes a new “ultra FPS” (frames Per Second) lidar technology concept. Through rotating-mirror technology, the objects within the lidar’s ROI (Region of Interest) will acquire a 20Hz point cloud data when the whole system is working at a frame rate of 10Hz. The new ROI point cloud density is hence increased to 144 lines equivalent at 0.1 second without the need for extra laser transmitters.
Since its founding in 2016, Livox has focused its R&D efforts on enabling cost-efficient mass production of lidar solutions, aiming to overcome the three bottlenecks in the lidar industry: price, scalability and reliability. Livox’s automotive-grade solution boasts a range of industry-leading standards in terms of detection range, FOV (Field of View) and point cloud density.
Xpeng has developed an autonomous driving architecture with multiple modules for mass production vehicles: perception, behavior planning, behavior/motion prediction and map fusion. Xpeng’s autonomous driving platform already integrates camera, millimeter-wave radar, ultrasonic and other sensors. The addition of lidar will not only provide a greater level of safety redundancy, by allowing more accurate imaging of the road situation, but will also enhance target detection, measurement accuracy, performance in low ambient light and other challenging perception conditions, and will significantly improve the overall performance of NGP (Navigation Guided Pilot), AEB (Autonomous Emergency Braking) and other functions.
