Analysis of key technologies in automotive electronic and electrical architecture software

1. Technology trends of automotive electronic and electrical architecture and in-vehicle computing platforms

The automotive electronic and electrical architecture is changing from the traditional distributed architecture to the domain architecture and central computing architecture. The in-vehicle control system tends to form a unified architecture standard and a universal software and hardware platform, and various control functions are gradually evolving into various applications under a unified platform. There are four key trends in its technological evolution: computing centralization, software and hardware decoupling, platform standardization, and functional development ecology. Intelligence and networking have jointly promoted the transformation of automotive electronic and electrical architecture. On the one hand, it is the optimization of the in-vehicle network topology and the activation of real-time, high-speed networks. On the other hand, the functions of ECU (electronic control unit) are further integrated into the domain controller and even the central computing unit.

Figure 1 Schematic diagram of the transformation trend of automotive electronic and electrical architecture

The underlying hardware of automotive electronics is no longer a single chip that implements a single function and provides simple logical calculations, but needs to provide more powerful computing support; software is no longer developed based on a fixed hardware, but must have characteristics such as portability, iterativeness and scalability. The original R&D organization of automobiles based on ECUs will change to form a new R&D organizational form of general hardware platforms, basic software platforms and various application software.

Figure 2 Schematic diagram of the trend of changes in the development of automotive electronic and electrical architecture

The consensus on the overall technical development trend of the vehicle computing platform is that the software is upgradeable, and software can be reused across models, software, and car companies; the hardware is scalable and replaceable, and the sensors are plug-and-play. Under the general trend of software and hardware decoupling, the software and hardware iteration cycle is accelerated to achieve a scalable vehicle computing platform. Neusoft Group proposed the idea of ​​a pluggable hardware architecture (as shown in Figure 3), using modular architecture as the entry point, and realizing fault detection, isolation, and blocking through chip redundancy and disaster tolerance, as well as the setting of computing power flow based on service containers, distributed computing, and service-oriented architecture. Huawei starts with driving software-defined cars through computing and communication architecture (as shown in Figure 4), building a trusted system, optimizing the cost of a single vehicle, reducing the vehicle development cycle based on a scalable architecture, smoothly promoting intelligent driving, and turning cars into platforms that can continuously create value.

Figure 3 Schematic diagram of Neusoft's automotive in-vehicle computing platform solution

Figure 4 Schematic diagram of Huawei's automotive in-vehicle computing platform solution

2. System level software

Key technology analysis

System software, or operating system, is the bottom layer for managing and controlling the hardware and software resources of smart cars, providing operating environment, operating mechanism, communication mechanism, and security mechanism, etc. Currently, vehicle-mounted operating systems can be divided into four levels: basic operating system, customized operating system, ROM operating system, and middleware.

The basic operating system includes the system kernel, underlying drivers, etc., which provide the most basic functions of the operating system, are responsible for managing the system's processes, memory, device drivers, files and network systems, and determine the system's performance and stability; the underlying operating system is currently an open source framework, which is not affected by copyright and intellectual property rights, and is generally not within the scope of technology that enterprises consider developing. The customized operating system is deeply customized on top of the basic operating system, such as modifying the kernel, hardware drivers, runtime environment, application framework, etc., and is an independent operating system independently developed. ROM is based on the modified system services and system UI of the release version. ROM-type automotive operating systems are limited customized development based on basic operating systems such as Linux or Android, which does not involve changes to the system kernel, and generally only modifies and updates the applications that come with the operating system. Most OEMs generally choose to develop ROM-type operating systems. Foreign OEMs mostly use Linux as the underlying operating system, while domestic OEMs prefer the Android application ecosystem. Middleware is software between applications and operating systems, which realizes commonalities and problems such as software interconnection and interoperability in heterogeneous network environments, provides standard interfaces and protocols, and has high portability. At present, domestic manufacturers are relatively advanced in the development of middleware, and are committed to providing transitional solutions in the transformation of E/E electronic architecture.

Figure 5 Schematic diagram of system layer software architecture

3. Autonomous driving software

Key technology analysis

The basic process of autonomous driving is divided into three parts: perception, decision-making, and control. Its key technology is the software algorithm and model of autonomous driving. By integrating the data of various sensors, different algorithms and supporting software can calculate the required autonomous driving solution. Environmental perception in autonomous driving refers to the ability to understand the scene of the environment, such as the type of obstacles, road signs and markings, detection of driving vehicles, classification of traffic information and other data. Positioning is the post-processing of the perception results, which helps the vehicle understand its position relative to the environment through the positioning function. Environmental perception requires a large amount of surrounding environmental information to be obtained through multiple sensors to ensure the correct understanding of the vehicle's surrounding environment, and to make corresponding plans and decisions based on this. There are currently two mainstream technical routes. One is a multi-sensor fusion solution dominated by cameras, represented by Tesla; the other is a technical solution dominated by laser radar and assisted by other sensors, represented by Google and Baidu. Decision-making is based on the cognitive situation map of the driving scene and makes task decisions according to driving needs. Then, under the premise of avoiding existing obstacles, it can plan multiple safe paths between two points through some specific constraints, and select the best path among these paths to decide the vehicle's driving trajectory. The execution system executes driving instructions and controls the vehicle status, such as the longitudinal control of the vehicle and the driving and braking control of the vehicle. The lateral control is the adjustment of the steering wheel angle and the control of the tire force. By achieving longitudinal and lateral automatic control, the vehicle operation can be automatically controlled according to given goals and constraints.

Figure 6: Key technology architecture of autonomous driving

4. Smart cockpit software

Key technologies

The analysis of the smart cockpit mainly covers the innovation and linkage in the cockpit interior and cockpit electronics fields, and the human-machine interaction (HMI) system built from the perspective of consumer application scenarios. The smart cockpit collects data and uploads it to the cloud for processing and calculation, so as to adapt resources in the most effective way and increase the safety, entertainment and practicality in the cockpit. The current smart cockpit mainly meets the functional requirements of the cockpit. On the original basis, it integrates existing functions or scattered information to improve cockpit performance, improve human-machine interaction methods, and provide digital services. The future form of the smart cockpit is "smart mobile space". Under the premise of the high popularity of 5G and the Internet of Vehicles, the smart cockpit is integrated with high-level autonomous driving, and gradually evolves into a smart space integrating "home, entertainment, work, and social".

Figure 7 Analysis of key technologies of smart cockpit

5. Internet of Vehicles Software

Key technology analysis

The Internet of Vehicles is a large-scale system network based on the in-vehicle network, inter-vehicle network and in-vehicle mobile Internet, which conducts wireless communication and information exchange between "people-vehicle-road-cloud" according to the agreed communication protocol and data interaction standard. It is an integrated network that can realize intelligent traffic management, intelligent dynamic information services and intelligent vehicle control, and is a typical application of Internet of Things technology in the field of transportation system. At the networking level, it is divided into three levels according to the different contents of networking communication: networking auxiliary information interaction, networking collaborative perception, and networking collaborative decision-making and control. At present, the industry is in the stage of networking auxiliary information interaction, that is, based on vehicle-road and vehicle-background communication, the acquisition of auxiliary information such as navigation and the upload of data such as vehicle driving and driver operation are realized. Therefore, at this stage, the Internet of Vehicles mainly refers to information services derived from networking auxiliary information interaction technology, such as navigation, entertainment, rescue, etc., but the generalized Internet of Vehicles, in addition to information services, also includes V2X-related technologies and services for realizing functions such as networking collaborative perception and control.