Autonomous vehicle startup announces availability of driving sensing solutions for purchase to OEMs
DeepRoute has released for sale its autonomous vehicle sensing solution, DeepRoute-Sense. DeepRoute is a CES 2020 Innovation Award Honoree.
DeepRoute-Sense is an L4-level self-driving full-stalk solution that aims to enable the autonomous vehicle industry to achieve quicker sensor deployment. It includes a sleek, lightweight set-top box and sensor-fusion calibration service.
The roof box consists of GNSS, eight vehicle cameras, three lidars and a series of other sensors to help correspondence and data synchronization between the controllers.
DeepRoute has independently designed the mechanical structure of the roof box, which has outstanding diversion, heat dissipation and sealing functions. The set-top box is lightweight, small and with high impact resistance. With four tripods, the roof box can be easily mounted to the roof of different vehicle models for sensor deployment.
“The team at DeepRoute has worked hard on the development of DeepRoute-Sense and we are excited to finally share our technology with the industry,” said Shuang Gao, COO of DeepRoute. “By bringing this to the market, we are hoping to fuel progress within the industry and bring full vehicle autonomy to be one step closer to the masses.”
The solution uses a DeepRoute-developed in-vehicle camera that features anti-glare, anti-ghosting and highly reliable signal detection. Compared with industrial cameras, the camera comes at a lower cost, but also offers a higher dynamic range.
The roof box has been developed to accommodate different and extreme weather conditions, whether it be during intense sunlight or nighttime high-beam illumination, and can stably handle the exposure and avoid overexposure.
The set-top box includes a sensor data-processing device, the ADS Synchronous Controller. It is pre-processed and fused with the massive data from GNSS, cameras, lidars and other sensors through high-precision time and space synchronization. The ADS Synchronous Controller also supports DNN and SLAM hardware acceleration, which greatly increases computational efficiency.