OTA cloud service
With the rapid development of new energy vehicles, under the trend of intelligence and networking, software-defined vehicles have become an industry knowledge, and OTA has become the core force of software-defined vehicles. Huawei released the OTA 3.0 vehicle-level upgrade scheme, which can provide a vehicle-level upgrade experience with zero brick experience.
Shorten the time-to-market cycle: the hardware is embedded and delivered, and the software is constantly iterated, which subverts the traditional mode of simultaneous delivery of software and hardware, and the time-to-market cycle of the whole vehicle is greatly shortened. Product evolution brings user experience improvement: Compared with traditional models, OTA users can continuously gain new functional experiences, including intelligent driving, intelligent vehicle control, power optimization and other fields, which will bring more intelligent experiences to car owners and enhance user stickiness.
It has brought great advantages to after-sales service: it has OTA capability, and can solve a large number of software problems that need to be repaired in 4S stores through remote upgrade, which greatly reduces the after-sales operation and maintenance cost.
Bring new business model: Compared with the traditional business model of one-time sales, users can pay for new software functions and hardware upgrades after buying a car, which will bring continuous incremental benefits to car companies.
The development of OTA has gone through three stages. From the market point of view, at present, most of the new forces ICV have taken the lead in entering the OTA 3.0 stage, and maintained a high iteration speed in the fields of ADAS/ADS, vehicle control, cockpit and power. Three core issues that OTA3.0 needs to pay attention to: vehicle version management and quality care, OTA safety and reliability, and user experience.
The first stage: writing with a brush near the end. The owner drove the car to the 4S shop and updated the software through the near-end diagnostic instrument. As early as 265438+the beginning of the 20th century, foreign car companies explored the remote upgrade of Tbox application for the first time, but it was limited to some basic network services, such as road rescue, ecall, remote query service and so on.
The second stage: OTA upgrade of parts. The upgrading form in this period is mainly the upgrading of a single component or a small number of components. In addition to the upgrading of basic network services and cockpit entertainment applications, some ECU FOTA based on CAN bus are gradually opening up. Due to the high security sensitivity of these ecus, OTA security issues have attracted attention for the first time.
The third stage: vehicle-level upgrade. This is an upgrading revolution of the intelligent experience of the whole vehicle, benefiting from the development of 4G/5G networks and intelligent parts of the whole vehicle. The functions of vehicle-level applications, such as intelligent driving and scene-based vehicle remote control, include cross-domain multi-ECU cooperation, so vehicle upgrade has become a necessary scenario for OTA, which requires the OTA system to support the whole process and whole scene upgrade capability from R&D to commercialization.
Vehicle version management is a systematic problem, how to ensure the compatibility and consistency of vehicle parts versions. At present, the number of lines of on-board software code reaches 1 100 million lines, and the complexity of multi-component collaboration is high. After-sale vehicles have a variety of versions of accessories, resulting in different vehicles. How to ensure the compatibility and consistency of vehicle parts versions, how to ensure the quality of versions and prevent bugs, especially those that affect driving safety, is a huge project. Taking 20 parts combinations as an example, in the life cycle of a vehicle, the number of versions caused by changes in functions and problems may reach thousands. In the era of vehicle intelligent experience, OTA is becoming more and more popular, and efficient version management has become a key challenge. OTA platform which can provide effective software version management capability has more advantages.
Intelligent driving, vehicle control and power domain upgrade have become the norm, and how to ensure the safety and reliability of the vehicle under OTA is extremely critical. OTA upgrade will lead to brick replacement, which will touch the red line of user experience, and complaints and even public opinion cases caused by brick replacement in the industry will emerge one after another. Huawei systematically analyzed the brick changes caused by OTA. Control loopholes, unavailability of core components, lack of remote maintenance and software bugs in the process of vehicle upgrading have become the biggest incentives for brick replacement. How to predict all faults that may lead to brick upgrading in the design stage is a systematic problem. It is necessary to analyze and simulate the potential faults in all scenarios and formulate attenuation measures to prevent them. Huawei mobile phones supported OTA many years ago, and systematically analyzed the failure modes of mobile phone OTA failures, which also ensured billions of reliable upgrades. Experience is worth learning.
Similar to mobile phone upgrade, it also faces user experience problems such as traffic consumption, upgrade time and operation interaction in OTA process.
Huawei proposes to take OTA as the core service, and OTA3.0 era needs to have three major capabilities: to get through the whole process, cover the whole scene, and be safe and reliable;
Huawei OTA solution is a vehicle-level remote upgrade service for Huawei HI automotive solutions. The upgrade of natural adaptation HI solution can cover 45+ components such as intelligent driving, intelligent cockpit, intelligent electric, intelligent network connection and intelligent vehicle control; At the same time, Huawei's open car-end upgrade service architecture supports the upgrade of car companies' own parts and three-party parts, and can support the writing specifications of standard parts of car companies; At the application level, it supports upgrade functions, including ADAS/ADS applications, algorithms, HarmonyOS operating system and HarmonyOS local applications.
In the era of smart cars, various car companies have continuously developed software ecology, constructed and optimized software management processes, and have professional opinions and thoughts on the reliability of vehicle upgrades, and also accumulated rich experience. Huawei provides an OTA service platform, and hopes to continue to explore OTA software ecology, management process, reliability, operation and maintenance with various car companies and ecological partners to create competitive OTA services.
VHR
At the same time, with the development of vehicle intelligence, the digital proportion of each body part is also constantly improving: the magnitude of vehicle software code is constantly improving, and now the amount of software code is 10 times that of ten years ago, and the amount of software code will be further increased to 300-500 million lines in the future. Every system and component of the vehicle reports various status signals, logs and alarms every day. At present, the data volume of a bicycle is on the order of 150-200M per day, and it will be further improved with the increase of sampling frequency and component complexity in the future.
Referring to the experience of ICT industry, Huawei put forward the concept of VHR. VHR stands for vehicle history, which is a data-driven concept based on the whole life cycle. Its purpose is to realize the visibility, maintainability, user care and efficient operation of vehicles on the basis of a large number of data. These works will greatly enhance the experience of car owners and user stickiness in the future, and also bring important value and benefits to car companies.
VHR covers data collection, data management, data analysis, vehicle state visualization, vehicle fault diagnosis, trend analysis, prediction and improvement. It is a closed-loop system from vehicle to vehicle departure.
There will be many application scenarios based on VHR in the future. Here are five main scenarios:
Scene 1: Digital twins of the vehicle
Visualization of vehicle core domain and core components, such as motors and batteries in power domain, digital chassis in chassis domain, and sensors (radar, camera, etc.) for automatic driving or advanced assisted driving. ), in-vehicle intelligent driving computing platform, intelligent cockpit, etc. It can be realized by digital pairing, especially for the component system which is particularly important for life safety, and it can know the running state, key parameters and indicators of these components in real time, which is very important for product improvement, defect discovery, performance improvement and fault demarcation of smart cars.
Digital pairing is definitely not a simple perspective. Only by deeply understanding the key structure and performance parameters of each field and each system can we build a truly effective digital pairing system. Moreover, this system has high requirements for the platform's big data and AI capabilities.
Scenario 2: Remote Diagnosis
The traditional way to deal with car faults is to go to the maintenance center. For traditional cars, this model is not a big problem, but smart cars have several core changes. One is that the power system has changed from a fuel engine to a power battery and motor. Whether the power battery is faulty or its performance changes with time, so it is very necessary to monitor it for a long time and find out the potential risks in advance (this will be mentioned in the following three power cloud services).
Another change is that in the future, in intelligent driving and advanced assisted driving scenarios, the requirements for reliability will be very high, so it is very important to predict and remotely diagnose the faults of the core components of the system.
Therefore, from the dimensions of vehicles, fields and components, diagnosis is also divided into three levels. According to the opinions of industry experts and OEMs, Huawei refined some core scenarios, such as vehicle collapse, thermal runaway, collision, OTA upgrade failure, and insufficient braking force. By constructing fault tree or AI learning, the ability of remote diagnosis can be improved in a targeted way.
Scenario 3: Smart tariffs
I believe everyone has had such an experience. When you have questions about the goods or services you buy, you will call customer service. Now most of you will use online customer service on the Web or in the APP. If customer service repeatedly asks for some basic information and information, when customer service A transfers it to customer service B, you may be dissatisfied with the repeated description of the questions and basic information, which will reduce your service experience. Therefore, it is very important to build some basic capabilities in the customer service center, such as the basic information of the vehicle, the past maintenance history of the vehicle after it goes offline, the state of the vehicle, and suggestions for solving some basic problems. This is very important to enhance the perception of car owners.
Through VHR, you can know more about the car than the owner, and always escort the users through intelligent duty, so that the user's stickiness will be greatly enhanced. Under the same product strength, such a service experience will definitely leave a deep impression on users.
Scenario 4: Quality Prediction
China's automobile recall system has played an increasingly important role in improving the product quality of enterprises and protecting consumers' rights and interests.
For car companies, identifying potential risks and defects as early as possible will play an important role in greatly reducing quality costs, improving product and service quality and protecting brand value. Therefore, it will be a long-term and valuable work to get through the data of the whole life cycle from production line (pregnancy in October), vehicle use process (growth) and vehicle delisting, and build a quality analysis and prediction model based on these long-term data.
The data shows that in 2020, China recalled 6.782 million vehicles, including 45 new energy vehicles, involving 357,000 vehicles. As long as it is slightly improved, it will bring great benefits.
Scene 5: Vehicle Portrait
Through the integration and docking of MES, DMS, CRM, warranty, marketing system, OTA and other business systems, the domain model is constructed, the data is mined and correlated, and the user tag is formed. Through user portraits, more direct business innovation and data realization can be realized.
Cloud heterogeneous computing resources support model training calculation, fully integrate the capabilities of NPU, GPU and CPU, realize the integrated application scheduling of the underlying heterogeneous computing resources, and improve the efficiency of model training and execution.
Sensitive personal data end-side training (privacy protection), multi-user feature parameter cloud * * * supports joint learning (model accuracy), and increases noise data through differential privacy to improve security.
At the end side, high computing power components such as VDC, CDC and MDC are used, and anomaly detection models are gathered at the end side for anomaly detection and rapid reasoning. Combined with cloud big data model and data supplement provided by suppliers, fault labels are improved, and the accuracy and real-time performance of problem identification are improved.
The data-based VHR system needs to be jointly built by the whole industry, including car-side capabilities, cloud capabilities, upper-level scenarios and so on. On the one hand, continuously improve the car experience of car owners and truly enjoy the value brought by smart cars. On the other hand, it can also create more value and benefits for car companies and achieve a win-win situation. Huawei hopes to contribute its strength in this system, give full play to its advantages in data platform and AI capabilities, and combine the profound insights of industry leaders and experts to do a good job in VHR system.
Three-point cloud service
Driven by national policies and other factors, today's car electrification is unstoppable. 202 1 As of July, the national sales of new energy vehicles exceeded12.29 million, up by 2 10.2% year-on-year. New energy vehicles have developed rapidly, but at present, the spontaneous combustion of new energy vehicles has not been effectively controlled. Accidents 124 were reported in 2020. By September 2026, 5438+0, there were 224 car burning accidents exposed by the media, and consumers began to shift from mileage anxiety to safety concerns.
There are many and complicated reasons for the thermal runaway accident of electric vehicles: internal short circuit caused by battery technological defects, electrical abuse, thermal abuse and mechanical abuse leads to local heating, and excessive charging, low-temperature charging and fast charging lead to a large amount of lithium precipitation and reduce thermal stability, all of which may lead to thermal runaway. Among them, the safety boundary of thermal runaway caused by internal short circuit is dynamic, and whether internal short circuit leads to thermal runaway needs more judging factors, such as short circuit type, short circuit internal resistance, SOC and so on. Therefore, based on electrochemical mechanism and machine learning technology, Huawei has built a more complex relationship model to realize accurate early warning of power battery safety of new energy vehicles.
Based on Huawei VHR data service base and battery pack simulation system, eight applications of power battery are constructed, including power battery thermal runaway warning, battery fault detection, battery health SOH evaluation, battery remaining life RUL prediction and other applications, to escort new energy vehicles.
In the aspect of battery safety warning, the recall rate of power battery fault detection and thermal runaway warning can reach above 80%, and the false alarm rate is guaranteed to be in a low range. In the aspect of battery health assessment (SOH), Huawei's cloud-based SOH prediction model can achieve a life cycle estimation error of less than 3% and a residual cycle prediction error of less than 10%, and can achieve a battery problem traceability of 100% based on cloud battery life cycle data.
Huawei builds a large number of sample data through the power battery hardware-in-the-loop simulation system, and combines the analysis of real vehicle sample data to effectively overcome the problem of small sample data. At present, Huawei has built fault sample data covering ternary lithium batteries and lithium iron phosphate batteries, and built a 15+ power battery thermal runaway feature engineering library.
Huawei adopts the domain adaptive algorithm migration algorithm model. When the algorithm model is migrated from the simulation environment to the real vehicle, or in different material systems and formulas, the accuracy of the algorithm remains stable.