Key design elements for intelligent driving: integrated positioning
Latest update time：2021-05-23
Reads：
Cars will face unprecedented changes in the field of autonomous driving. This change is not only closely related to the functional updates required during driving, but also includes major changes in the architectural design of the entire driving system development process.
Among them, high-precision maps, as an important aspect of positioning and navigation, will also undergo major design changes.
This is mainly reflected in the following important aspects:
1) The deployment of roadside maps requires more strong support from government infrastructure departments, including V2I-like Internet of Vehicles systems, providing V2X scene rendering SDK, supporting custom rendering styles, and realizing personalized vehicle-side V2X applications. Using the roadside information channel (PC5) can solve the problem of the impact of blind spots in cellular network coverage on map services.
2) The real-time distribution of maps has also been upgraded from the original second level to the millisecond level. High-precision map update and distribution capabilities tend to be deployed at the edge to achieve real-time map updates and millisecond-level data distribution services, aiming to improve the real-time nature of data transmission. .
3) The real-time high-precision map platform in the autonomous driving system is gradually transitioning from the original distributed system to a centralized system. The high-precision map engine can provide real-time high-precision map application services for autonomous driving systems and assist autonomous driving vehicle applications.
4) Maps need to establish more self-learning mechanisms, including establishing self-learning maps in shadow mode, and constantly updating the map's understanding and learning of the environment.
5) The autonomous driving function is redefined by the optimized map. This process is actually a process of continuously iterating the autonomous driving function based on the map.
6)
Provide diversified basic map services
In the future, high-precision maps will be continuously optimized and updated to provide corresponding management services and technical advantages, including complete function iteration, system data security, multi-type data support, multi-terminal authorization support, high-quality service and operation and maintenance system, and data online Upgrade services in several aspects.
Among them, the functional improvement indicators include: data distribution, collection, subscription, notification functions, dynamic data online compilation, visual online display editing, Campaign, Map Learning crowdsourcing updates, data simulation platform, etc.
Multi-data support includes: versioned static layers, effective dynamic data support, streaming data support, OSO custom data support, incremental data release and update, and Map data integrity assurance.
System data security includes: complete network security configuration, firewall, VPC, multi-level system permission management, users, roles, permissions, resources, https two-way encryption, data signature, and HSM device integration.
Multi-terminal permission support includes: OAuth2-based authorization management, massive concurrent responses, Auto Scaling dynamic scaling, complete API SDK development kit support, 2V terminal development kit and function integration, OTA updates, and VCDN to ensure data security and speed.
7) V2X high-precision map service supporting edge computing
In the future, the development of autonomous driving will mainly move towards two major directions: intelligence and networking. The focus is on the gradual transition from vehicle-end intelligence to road-end and even cloud-end intelligence. For the deployment of high-precision maps, the most important points are the series of upgrades and changes that have occurred on the cloud, roadside and vehicle side. Among them, cloud changes mainly involve high-precision map services, sharded data aggregation, data edges, and data sharding. The roadside end mainly involves several aspects such as sharded map crowdsourcing update, map version management, map subcontracting, roadside dynamic information optimization, and map message services. When applied to the car, it requires subcontracted data fusion, V2X scene restoration, and high-precision map engines to be updated accordingly.
The above iterative update process of the map can be applied to realize L4/L5 level autonomous driving functions and generate relevant robot control modes. It can also be used in the realization of commercial vehicles to ultimately achieve autonomous driving and even remote driving.
High-precision fusion positioning solution for mass production
Obviously, in order to achieve precise positioning and continuously extend and improve the functional performance of high-precision maps, it must be obtained by continuously optimizing its own integrated positioning solution. This process involves two main software algorithms. The first is to perform dynamic optimal estimation of vehicle pose through full-state extended Kalman filtering; the second is to use visual sensors to obtain semantic information of the road environment and obtain precise positions through precise map matching algorithms. In addition, there is a need to improve economy, fit and overall performance. By choosing to configure industrial-grade vehicle-mounted terminal RTK: using high-performance industrial-grade 32-bit processor, built-in high-precision RTK board; establishing a channel with Qianxun platform through 3G/4G/5G, sending GGA information to the differential server, and receiving differential signals at the same time After receiving the information, the precise location information is output through RS232.
By selecting consumer-grade sensors, that is, sensors that have been installed on the vehicle (such as cameras, lidar and radar information suitable for intelligent driving systems), they are used for positioning fusion (such as visual SLAM, laser SLAM) to improve positioning performance. For high adaptability solutions, we mainly use a unique hardware adaptation layer (such as an independent domain controller system) and a software adaptation layer (such as a standard C language interface) to avoid platform dependence. High-performance requirements mainly involve the output of requirement results for both horizontal and vertical positioning. Generally, the horizontal positioning error requirement is 20cm, the longitudinal positioning error requirement is 1m, and the heading angle positioning error is 0.5°. At the same time, the signal loss rate of GNSS in longitudinal positioning error should be less than 0.3%. In addition, it also needs to support 1000Hz external IMU and 50Hz frequency camera input solution.
In addition to the overall map architecture settings, its data release mode also requires minimum traffic costs to complete high-precision map updates and support streaming incremental release. This process includes streaming publishing an appropriate amount of tiles, incrementally updating data, browsing rich data sets, and distinguishing directories, levels, and tiles to enable hierarchical queries. Finally, retrieve cloud historical data on demand and trace back to any version of information at any time.
Map distribution and map packaging
The most important process of high-precision maps involves the collection and distribution of map crowdsourcing. Regarding the collection of crowdsourced map data, it can actually be understood that the road data collected by users through the self-driving vehicle's own sensors or other low-cost sensor hardware is transmitted to the cloud for data fusion, and the data is improved through data aggregation. accuracy to complete the production of high-precision maps. The entire crowdsourcing process actually includes physical sensor reporting, map scene matching, scene clustering, change detection and updating.
Where will the new architecture of autonomous driving developed based on maps go?
The current high-precision map architecture of the autonomous driving system is still oriented to a distributed approach. Its key concerns include map crowdsourcing collection, the map box's analysis of the original information of the high-precision map, and how the map fuses input data from other sensors. Let us note here that the future autonomous driving system architecture will continue to evolve from distributed development methods to centralized ones. The centralized approach can be viewed in three or two steps:
Step1: Fully centralized control solution for intelligent driving domain
That is, the intelligent driving ADS and intelligent parking AVP systems are fully centralized controlled, and a central pre-processing device is used to integrate, predict, plan and other processing methods for the information required to be processed in the two systems. The processing methods of all sensing and data units related to smart driving and smart parking (high-precision maps, lidar, fully distributed cameras, millimeter wave radar, etc.) will be integrated into the central domain control unit accordingly.
Step2: Fully centralized control solution for the intelligent driving domain and intelligent cockpit domain
This approach is the second stage to achieve a fully centralized distribution approach, which involves the development of all functions covered by the intelligent driving domain controller (such as autonomous driving and automatic parking) and the development of all functions covered by the intelligent cockpit domain (including driver Monitoring DMS, audio-visual entertainment system iHU, instrument display system IP) for integrated coverage.
Step3: Fully centralized control solution for the intelligent vehicle domain
Here is a fully integrated control method that includes intelligent driving, intelligent cockpit and intelligent chassis domains. That is, the three main functions are integrated into the vehicle central control unit, and the later processing of this data will create more performance (computing power, bandwidth, storage, etc.) requirements for the domain controller.
The high-precision map positioning development we are concerned about here will be more oriented towards centralized design methods in the future. We will elaborate on this.
The figure above shows the architectural development trend of high-precision maps in future autonomous driving system control. In the future, autonomous driving systems will strive to integrate the sensing unit, decision-making unit, and map positioning unit into the central domain control unit, aiming to reduce the dependence on high-precision map boxes from the bottom up. The design of its domain controller fully considers the full integration of AI computing chip SOC, logic computing chip MCU, and high-precision map box.
The figure above shows the corresponding high-precision map sensor data collection, data learning, AI training, high-precision map services, simulation and other services under the entire cloud control logic. At the same time, during the movement and verification process of the vehicle, physical sensing, Dynamic data sensing, map target sensing, positioning, path planning and other content continuously update the map data and upload it OTA to the cloud to update the overall crowdsourced data.
The previous article has described the process of how high-precision map data generates relevant data that can be processed by the autonomous driving controller. We know that the original data processed by high-precision maps is EHP data. The data actually contains the following main data support:
1: Received external GPS location information;
2: Match the location information to the map;
3: Establish road network topology information;
4: Send data through CAN;
5: Fusion of partial navigation data;
This data is generally directly processed from the HDMap sensing end through Gigabit Ethernet and then input to the high-precision map central processing unit. This central processing unit is called the "high-precision map box". Through further processing of the data through the map box (we will explain this actual processing process in detail in a subsequent article), it can be converted into EHR (actually CanFD) data that can be processed by the autonomous driving controller.
For the next generation of autonomous driving systems, we are committed to integrating high-precision map information into the autonomous driving domain controller for overall processing. This process means that our autonomous driving domain controller needs to receive all the data from the map box. analysis work, then we need to focus on the following points:
1) Can the AI chip of the autonomous driving domain controller process all sensor data required for high-precision maps?
2) Does the logic operation unit of the high-precision positioning map have enough computing power to perform sensor data information fusion?
3) Does the entire underlying operating system meet functional safety requirements?
4) What connection method is used between the AI chip and the logic chip to ensure the reliability of data transmission, Ethernet or CanFD?
In order to answer the above questions, we need to analyze the way the controller processes high-precision map data as shown in the figure below.
As the AI chip of the autonomous driving system, SOC is mainly responsible for the basic processing of sensor data in future high-precision map data processing, including camera data, lidar data, millimeter wave data, etc. In addition to basic data point cloud fusion and clustering, the applied processing methods also include commonly used deep learning algorithms, and ARM cores are generally used for central computing processing.
As the logical operation unit of the autonomous driving domain controller, the MCU will subsequently undertake all the logical calculations required by the original high-precision map box. Including front-end vector aggregation, sensor fusion positioning, building road network maps, and most importantly, replacing the original map box function to convert EHP information into EHR signals (how the central processing unit MCU can effectively convert EHP information into EHR information Will be detailed in a later article) and perform effective signal transmission through Can lines. Finally, AutoBox, a logical operation unit, is used for path planning, decision-making control and other operations.
Summarize
Future autonomous driving will tend to integrate all data information processed by high-precision maps from the original map box into the autonomous driving domain controller, aiming to establish a true central processing integration with the entire vehicle domain controller as the integrated unit. . This method can not only save more computing resources, but also enable the AI data processing algorithm to be better applied to high-precision positioning, ensuring the consistency of the two's understanding of the environment. We need to pay more attention to the important direction of high-precision sensor data integration in the future, and put more effort into chip computing power, interface design, bandwidth design and functional safety design.
Submission cooperation: 18918250345 (WeChat)
京公网安备 11010802033920号