On April 18th, Jidu once again showcased the progress of its first car robot.
Unfortunately, Jidu did not disclose the “ultimate form” of its first car robot at today’s communication meeting. Instead, they revealed some details about the car robot. We can see that Jidu’s car robot uses a lifting design for its two Velodyne AT128 LiDARs, which are located on the front trunk lid. This design is currently uncommon. Additionally, the first model will also feature scissor doors.
At the same time, Wang Weibao, the head of Jidu’s intelligent driving department, shared some recent developments of car robots in the field of intelligent driving. Currently, Jidu’s SIMUCar, which is the software simulation car of Jidu’s car, has been upgraded to version 2.0. The LiDAR sensor has been added to the automatic driving system and has been integrated with JET 1.0, Jidu’s self-developed high-level automatic driving intelligent architecture.
Wang Weibao also explained the benefits of placing the LiDAR sensor on the front trunk lid. Firstly, the dual forward-facing LiDARs can provide a better field of view angle, increasing the field of view angle from 120 degrees of a single LiDAR to 180 degrees. Secondly, the recognition accuracy of key areas in front of the vehicle is higher. Within the 60-degree field of view area in front of the vehicle, the dual LiDAR can achieve double overlap, providing more point cloud data on the target object and higher recognition accuracy. In addition, the dual LiDAR can also achieve redundancy.
However, Li Xiang, the CEO of Li Xiang Automobile, does not fully agree with this design, and re-posted on Weibo that “placing one LiDAR on the roof is no different in performance from placing two LiDARs on the hood or bumper. Even the single LiDAR on the roof is stronger in performance. However, placing the LiDAR on the roof is optimal for pedestrian collision, repair costs, and vibration control (connected to the main body of the car). In addition, the roof scheme has the best sensory effect.” Therefore, the competition of LiDAR positioning is not only in technical performance but also in safety positioning, and each company has its own perspective.
Nowadays, as more and more sensors are being developed and chip computing power is becoming larger, everyone is working hard on sensor fusion. However, Jidu has announced that its SIMUCar has already achieved “dual-domain integration” and redundancy. This means that the two sets of sensors installed on Jidu are not fused together to work together, and both are completely independent systems. This scheme is conceptually very similar to the truly redundant system of Mobileye.
Finally, we have summarized some of the progress made by Jidu Auto since its establishment:
- On March 2, 2021, Jidu Auto Co., Ltd., a joint venture between Baidu and Geely, was officially established, with Xia Yiping as its legal representative;
- On June 1, 2021, Jidu announced that it would jointly develop an L4-level intelligent architecture – JET with Geely, and based on the JET architecture, Jidu will develop a software architecture based on this SOA, and cooperatively develop it with Apollo and Geely.
- On November 2, 2021, Jidu Auto announced to the outside world the SIMUCar, which stands for Software- Simulation Vehicle，preliminary to the development of autonomous driving functions and digital intelligent cockpits.
- On November 16, 2021, Jidu Auto confirmed that Zhu Jiang, formerly vice president of NIO, would officially join Jidu Auto as Vice President of User Development and Operations and Vice President.
- On December 1, 2021, Jidu Auto, together with Baidu and Qualcomm, announced that Jidu Auto’s first mass-produced vehicle would be the first to launch in China with the Qualcomm 4th Generation Snapdragon Automotive Digital Cockpit Platform 8295 chip.
Jidu’s first car robot concept car was originally scheduled to be unveiled at the Beijing Auto Show in April this year, but due to the pandemic, the auto show has been postponed. In addition, Jidu’s first car robot will be mass-produced and delivered in 2023.
This article is a translation by ChatGPT of a Chinese report from 42HOW. If you have any questions about it, please email bd@42how.com.