Skip to content

DJI, similar to Tesla, is addressing the ADAS conundrum using a vision-based strategy

Challenges in Balancing Expenses, Efficiency, and Safety: Advanced Driver Assistance System Creators grapple with the task of ensuring all three facets simultaneously.

DJI, similar to Tesla, is addressing the challenges of the Autonomous Driving Assistance System...
DJI, similar to Tesla, is addressing the challenges of the Autonomous Driving Assistance System (ADAS) trilemma by employing a vision-centered strategy.

DJI, similar to Tesla, is addressing the ADAS conundrum using a vision-based strategy

DJI, the world-renowned drone manufacturer, has ventured into the autonomous driving (ADAS) market, aiming to disrupt the industry with its cost-effective solutions. Founded by Frank Wang, the founder of DJI, DJI Automotive began developing its autonomous vehicle technology around 2021.

In a bid to enhance obstacle avoidance and bypass capabilities across various scenarios, including urban navigation, highways, and parking, DJI has omitted the need for high-precision maps. Instead, the company has upgraded its algorithms to include BEV models based on Transformer, Occupancy, and online construction of road topology.

DJI's autonomous driving solution, priced at around RMB 7,000, poses a competitive offering that could catalyze widespread ADAS adoption. This price point is significantly lower than premium ADAS-equipped car models developed by companies like Nio, Xpeng, Li Auto, Huawei, and Weixiaoli, which typically exceed RMB 200,000.

To achieve this affordable cost, DJI Automotive has replaced LiDAR sensors with a pair of front stereo cameras for perceiving the depth of road obstacles, four surround-view fisheye cameras, and a rear monocular camera. The company is using a Qualcomm 8650 chip for its autonomous driving solution, which offers better cost-effectiveness compared to the Nvidia Orin X.

As DJI prepares to commence mass production of its autonomous driving solution in the third quarter, the software's control is expected to be more mature, and the user experience improved. The company is also working on developing more advanced solutions aimed at achieving Level 3 autonomous driving, involving inertial navigation supported by three cameras and LiDAR assemblies.

Access to big data, alongside cooperation with automakers and issuing OTA updates, are crucial areas that DJI needs to work on. The development of ADAS faces a cost-performance-safety trilemma, making it difficult to affordably develop high-quality systems at scale.

In China, several companies are racing to expand the coverage of advanced driver assistance systems (ADAS) nationwide. Huawei has deployed its second-generation ADAS system that doesn't require high-precision maps. Xpeng Motors has introduced its intelligent driving solution to 243 cities, while Nio recruited 20,000 users for testing across 706 cities.

As DJI continues to innovate and improve its autonomous driving solutions, it is clear that the company is poised to make a significant impact in the industry. With its focus on cost-effectiveness and real-time road topology construction, DJI's autonomous driving solution could become a popular choice for consumers and automakers alike.

Read also:

Latest