On April 18, 2022, Jidu Automobile held an online media communication meeting. At the event, Jidu publicly released its first car robot concept car using two lidar solutions.
From the published product map, the design of the lidar at the front cover is clearly visible, which is located on the left and right sides of the front cover.
Jidu uses HESAI Technology's AT128 in this concept car, and the full name of the product is "AT128 Vehicle Specification Grade Hybrid Solid-State Long-Range LiDAR". According to the product description on the official website of Hesai, the AT128 is a product for the production vehicle of the advanced assisted driving frontload, which can hit more than 1.53 million points per second in the three-dimensional space to obtain excellent environmental perception ability, and can effectively perceive vehicles and pedestrians at 200 meters (@10% reflectivity).
At the event, Wang Weibao, head of Jidu Intelligent Driving, shared the latest progress of Jidu automotive robots. Recently, Jidu SIMUCar (Software Integration Mule Car) has been iteratively upgraded to version 2.0, and its automatic driving system has been added to the lidar sensor, and integrated with jet1.0 (electronic and electrical architecture + SOA), a high-level autonomous driving intelligent architecture developed by Jidu.
Set degree SIMUCar 2.0 form factor
As of now, Nvidia's Orin in-vehicle chip and Hesai Technology AT128 semi-solid-state lidar have been put on the car to participate in the test of the jidu automatic driving system. The hardware collocation of the first echelon represents that Jidu's high-end autonomous driving technology solution for mass production based on Baidu Apollo's unmanned autonomous driving capability has been fully exposed for the first time.
Wang Weibao said that in addition to the extra points in appearance, the dual lidar configuration brings more advantages in terms of safety. First, dual lidar's point cloud field of view is wider and safer. Compared to the 120-degree FOV (Field of view) of the single-radar scheme, the wider the field of view, the more you see), the front-end dual-radar scheme achieves horizontal 180-degree FOV coverage. In driving scenes such as "ghost probes" and left and right obstructions, the ability to recognize pedestrians or obstacles crossing the left and right is stronger.
Use the human eye as an example
Secondly, the dual lidar scheme is more accurate in identifying the key areas of the vehicle. In the area of the 60-degree FOV directly in front of the vehicle, the dual lidar can be doubled and overlapped, and there is more point cloud data on the target, and the recognition accuracy is higher. In terms of redundant safety for high-level autonomous driving, dual lidar can be safely redundant with each other, which is more reliable than single lidar solutions.
Baidu Apollo's technology has always focused on the pure visual technology route. The mass production ANP Pilot Assisted Driving and AVP Autonomous Parking announced in 2021 are based on lidar-free pure vision solutions. When Jidu signed a contract with Hesai Technology in August 2021, it was once interpreted by the outside world as a pure visual route for Baidu's automatic driving that may be shaken.
As we all know, the dispute between radar and vision on the path of autonomous driving technology is not only inconclusive, but also has a growing trend. But this does not prevent the set from advancing at its own pace.
In December 2021, the two-domain integration of SIMUCar automatic driving was achieved by relying on pure vision solutions. On the 25-kilometer test route, SIMUCar completed a complete test of automatic driving scenarios such as traffic light recognition, straight, right turn, protected left turn, and passing through t-shaped intersections and up and down ramps, at which stage lidar has not yet been tested.
Wang Weibao said at the event that at this stage, the pure vision and lidar autonomous driving scheme being tested and developed in the SIMUCar 2.0 stage is an independent dual system that can achieve "true redundancy" after mass production. The two sets of automatic driving solutions of Jidu are mutually backup and complement each other, and the two sets of solutions can not only drive themselves efficiently, but also coordinate efficiently.
Wang Weibao believes that the SIMUCar 1.0 stage, like a person's primary school stage, is a key period to lay a good foundation, focusing more on the basic capacity building of automatic driving, including the development of software architecture and related basic functional logic, and adapting to the chassis.
SIMUCar 2.0 is like the middle school stage, which is a period of gradual improvement and formation of basic capabilities, in this stage, the underlying architecture of SIMUCar has been upgraded to a high-level autonomous driving intelligent architecture JET (JIDU Evolving Technology) developed by jidu, including the electrical and electronic architecture EEA and the vehicle operating system SOA. At this time, the relevant domain controllers and sensors have gradually reached the mass production state, and some basic automatic driving capabilities, including ADAS functions, parking, intelligent interaction, etc., will gradually reach the mass production state.
The faster the run, the faster the set
Just over a year after its official establishment in March 2021, Jidu has grown from 15 people at the beginning to a team of more than 1,000 people. Although young, in terms of manufacturing industry topics, jidu can be said to have not slowed down.
First Qualcomm SA8295: On November 29, 2021, Jidu Automobile announced that it has joined forces with Baidu and Qualcomm to use the SA8295 chip of 5nm process on the first model of Jidu to be launched in 2023 on the 4th generation Snapdragon digital cockpit platform. As the world's first vehicle-grade 5nm chip, SA8295 can provide the strongest computing power, the strongest I/O capability and the strongest AI learning ability in the industry.
Dual-domain intelligent driving integration: In December 2021, Jidu SIMUCar successfully integrated high-speed and urban dual-domain intelligent driving, and initially had the ability to move freely in corresponding scenarios.
Nearly $400 million in financing: On January 26, 2022, Jidu announced that it had completed nearly $400 million of Series A financing, which was jointly increased by Baidu and strategic partner Geely. Turning back 10 months, Jidu has received more than $300 million in start-up capital since its inception.
Together with ZF: On February 22, 2022, Jidu and ZF formally signed a strategic agreement to work closely together in chassis components and control systems, electric drive systems, passive safety systems, etc., and jointly develop intelligent chassis technology.
Write at the end
Jidu originally planned to release its first automotive robot concept car in 2022, which was released at the Beijing Auto Show. Affected by the trend of the epidemic, the first large-scale brand launch event originally scheduled for April, ROBODAY, was also postponed with the 2022 Beijing Auto Show.
For Jidu, which has just been established for more than a year, the epidemic will not slow them down. The first mass-produced automotive robot, originally scheduled for delivery in 2023, is still on the schedule, and development and pre-development of subsequent models have also started. Jidu said that its second mass production car is expected to be released at the Guangzhou Auto Show at the end of 2022.
Finally, is it only our opinion that the design of the dual lidar scheme announced by Jidu this time looks a little bit like the armed to the teeth of the luxury car that Dr. Q built for 007?
In Die Another Day
Aston Martin Vanquish