Recently, I have been deeply thinking about some investment opportunities in the second half of smart cars, and I have done some combing on the correlation between new energy vehicles and smart cars on the market.
Here are some characteristics: in the past few years of new energy vehicles, Tesla and new forces (Weilai, Ideal and Xiaopeng) do have some big investment in intelligent and automatic assisted driving, but the A00 we see is to adopt a cost-first path to intelligence; and BYD and GAC Aeon and other traditional car brands, their electric vehicles, or in the past configuration around the traditional ADAS to do, switch to independent development software to achieve SOP and iterative maturity, A 2-3 year conversion period is required.
From the current point in time:
● In the next 2-3 years, we can see these backbone models gradually convert from the functions of L1-L2, which are basically provided by foreign parts, to the algorithm model of independent integration and domestic joint venture.
● With the increase in the number of new car manufacturers, the corresponding independent traditional car companies will also switch on the new platform, and from the perception level, they will also go up to a higher level in driving and parking intelligence.
Part 1
Automatic assisted driving configuration of Han EV and Qin DM-i
This part is that I carefully sorted out the perception and configuration of Han EV and DM-i cars, and in the publicity, BYD went to a name for The automatic assisted driving of Han EV called DiPilot (DiDAS+DiTrainer).
DiDAS features include AEB, Forward Collision Warning System, ACC, Traffic Congestion Assist, LCC, Lane Keeping System, BSD, Automatic Parking, Panoramic Imaging, Remote Driving, and more. The complete set of automatic driver assistance hardware combination consists of 3 mm wave radar (front one and rear two), 1 monocular camera, 12 ultrasonic radar, and 4 surround view cameras.
Figure 1: Bydir Han EV's autonomous driving hardware configuration
The Qin of the DM-i series has little difference in the entire driving configuration, that is, the parking system level and the previous BYD E2.0 system have been integrated for a while, and this piece is also integrated into the body system.
Figure 2: Qin DM-i's automatic driving hardware configuration
That is to say, in the powertrain update represented by the blade battery (EV and PHEV), BYD has a relatively high degree of self-research and completion in the integration of entertainment and body electronics; the resources allocated to automatic driving are relatively small. Therefore, we pressed BYD's automatic assisted driving, mainly around the mature foreign supplier L1-L2 way to achieve large-scale mass production.
Figure 3: The integration of the previous E2.0 platform in terms of electrical functions
Part 2
Next-generation E3.0 architecture
My understanding is that last year's E3.0 architecture came out, in terms of power and car control, the landing is relatively fast - wait for the dolphin to disassemble out, we can go to see; but in terms of intelligent driving, switching takes time.
At the end of last year, BYD and Momenta officially announced the establishment of an intelligent driving joint venture company Dipai Zhixing, from this import cycle, BYD may need to implant the algorithm into its own platform, due to the accumulation of previous intelligent driving data, iteration also needs a cycle, we can generally look forward to it according to about 2 years.
Figure 4:BYD's E3.0 intelligent architecture
The interesting thing is that BYD is at the basic and cooperative level in terms of intelligence, and the third generation of pure electric driverless delivery vehicle jointly developed and designed with Nuro is officially released, AND ISD is responsible for vehicle development, vehicle testing and manufacturing, while providing core first-level assembly components such as blade batteries, motors, electronic controls and human-computer interaction. Nuro offers technologies for autonomous driving, gateways, control modules and sensors.
Figure 5. Driverless car
Summary: I personally feel that intelligence, especially the change of automatic assisted driving, exists, some of the most conservative car companies are going in the direction of self-research, to make up for the overall shortcomings, with the iteration of 2-3 years, added value will rise, demand will also rise.