Source: Synthesized from the Web
Author: Ada
Internet of Things think tank collation and release
Guide
A few days ago, the group standard of "General Technical Conditions for Road Intersection Traffic Information Holographic Collection System" led by Huawei was officially released, which once again entered the public eye of "holographic intersection", a new favorite of intelligent transportation, and also opened a new battlefield for high-precision maps.
All along, the lagging transportation system has become more and more "bulky" against the background of the rapid development speed of the city, and to this day, the morning/evening peak congestion caused by unreasonable road layout and frequent accidents is still a pain that is difficult to get rid of in first-tier cities such as Beijing, Shanghai, Guangzhou, and Shenzhen. Especially after the large-scale landing of automatic driving has been put on the agenda, the complex road situation has become another mountain in front of the intelligent networked car.
Holographic intersection - perspective of intelligent traffic hazards
It is true that under the vigorous promotion of local governments, intelligent transportation systems have been deployed on a large scale in many cities in recent years, and global traffic management, intelligent traffic lights, signal control optimization, etc. have also begun to bear fruit, but it is obvious that they still cannot meet the travel needs of citizens. At the same time, we can still see the traffic police or auxiliary police at the intersection to guide the traffic, which also shows from the side that the reform in the traffic field still has a large space for development, at least it is not yet possible to achieve unmanned intelligent management.
On the one hand, the traditional intelligent traffic solutions in the face of complex intersections often rely too much on front-end cameras to capture pictures, videos, and then from the cloud for data processing, feedback to the control center, the process is lengthy, while traffic accidents are generally based on manual alarms, easy to cause "small accidents and large congestion" situation; on the other hand, the traditional signal control is "car lights", even if the optimized intersection signal control is only based on long-term data and the formation of prejudgment, can not be adjusted in real time according to the road conditions.
In recent years, with the continuous development of technologies such as edge computing, machine vision, high-precision positioning, digital twins, and holograms, people are also continuing to introduce new technologies and new ideas into the field of intelligent transportation in order to solve this social problem.
A few days ago, the group standard "General Technical Conditions for Road Intersection Traffic Information Collection System" jointly approved and compiled by Huawei, Beijing Tumeng Technology Co., Ltd., Beijing Chinasoft Zhengtong Information Technology Co., Ltd., Shenzhen Municipal Public Security Bureau Traffic Police Department, and Changzhou Municipal Public Security Bureau Traffic Police Detachment was officially released, which also made the "holographic intersection", a new favorite of intelligent transportation, once again enter the public vision, and can it really see the hidden dangers behind traffic congestion?
The holographic intersection solution adopts multi-directional lightning vision fitting technology, combined with high-precision map to present the intersection holographic perspective, through the automatic perception of traffic accidents and events, accurate data assists in determining responsibility, which can reduce the occurrence of accidents, provide accurate lane-level traffic data, and support intersection information control adaptive timing.
Industry experts said that holographic intersections serve future urban travel with six characteristics: global safety, scientific evaluation and fine governance, active service, new policing, data operation, and automatic driving.
In fact, as early as May 2020, Huawei released a holographic intersection solution based on its AI ultra-micro optical bayonet, millimeter wave radar, and intelligent transportation micro-edge ITS800, which uses multi-directional lightning fit technology to conduct data structured analysis of connected radar, video and other equipment, combined with high-precision maps to present a digital God perspective, which can achieve accurate portrayal of the running trends of motor vehicles, non-motorized vehicles, and pedestrians.
According to Huawei's official website, its holographic intersection solution can achieve:
Automatic perception of traffic accidents and events, accurate movement to assist in determining responsibility, reduce the occurrence of secondary accidents;
Provide accurate lane-level traffic data to support intersection information control adaptive timing;
Through real-time traffic heat map and road network travel rules, traffic hidden danger points are quickly discovered, the treatment cycle is shortened, and reasonable suggestions for traffic organization optimization are given according to data analysis.
It can be seen that Huawei's solution can realize the real-time digital twin of the operation of traffic intersections by converting the video of the four directions of the traditional intersection into a digital holographic perspective, combined with a high-precision map, and at the same time relying on its back-end algorithm, the intersection signal light control is upgraded from the traditional fixed mode of "car to see the lights" to the real-time mode of "lights to see the car", which is more flexible.
According to the author's understanding, after Huawei took the lead in launching holographic intersection solutions in 2020, Baidu and ZTE also launched smart intersection solutions in 2021. At the same time, on the demand side, since 2020, Shanghai, Zhuhai, Shaoxing and other places have also issued bidding for holographic intersection and smart intersection construction projects, with bids ranging from one million to tens of millions.
HD Map - Open the Holographic Intersection God Perspective
Looking at the holographic intersection solution, in addition to the front-end AI camera, millimeter wave radar, edge computing equipment and other hardware, the high-precision map, a technology with strong "automatic driving" attributes, is the key to opening the holographic perspective of real-world road conditions, which can more clearly depict the highway model on the one hand, and more accurately reflect the real-time road conditions of each lane on the other hand.
The so-called high-precision map refers to an electronic map with higher positioning accuracy and richer data dimensions, which can achieve accurate positioning at the lane level, which is the basis for the three-party collaboration of the car and the road, so it is also regarded as a new map data paradigm for self-driving cars.
HD map can be divided into two kinds, one is a static HD map, generally composed of three types of vector information containing semantic information, lane model, road components, road attributes, and feature layers for multi-sensor positioning; the other is a dynamic map, based on a static high-precision map, which mainly includes real-time dynamic information, covering traffic environment information, such as traffic lights, road congestion, construction, traffic control, weather conditions, etc.
Compared with the traditional navigation maps commonly used by people, high-precision maps mainly add a large amount of assisted driving related information, and greatly improve the map positioning accuracy and real-time data update capabilities.
Source: IoT Media
There is no doubt that the holographic intersection solution uses dynamic high-precision maps, through the combination of camera + radar, not only improve the positioning accuracy to lane-level information, but also can provide real-time traffic information, from the realization of a more realistic, detailed traffic intersection digital twin.
As a pathfinder for holographic intersection solutions, Huawei obtained the Grade A surveying and mapping qualification for navigation electronic map production in July 2019, and officially launched the commercial use of high-precision maps in 2021, launching a "high-precision positioning service", and users only need to pay 1 yuan per month to achieve sub-meter-level lane navigation services in some models.
It is reported that its high-precision mapping system is named Roadcode, which is divided into two parts - Roadcode HD and Roadcode RT. Roadcode HD is equivalent to a high-precision map drawn by a team of professional surveyors and mapping and is offline; Roadcode RT is equivalent to a self-learning map.
Of course, the players of the HD Map Track are not only Huawei, Tencent HD Map also delivered 358,000 kilometers of HD data across the country in 2021 as planned, and is expected to complete the full coverage of the urban two- to four-level roads in the top 100 cities in the country in the fourth quarter of this year; AutoNavi released the third-generation car navigation in January 2021, through the seamless switching of ordinary navigation maps and HD maps, to achieve "human-vehicle co-guidance", while Xiaopeng Automobile became the world's first enterprise equipped with AutoNavi's third-generation vehicle navigation...
Just in March this year, the semiconductor giant Nvidia also came to the high-precision map track to mix a foot, announced its high-precision map platform Drive Map, which will provide high-precision maps of 500,000 kilometers of roads around the world, with an accuracy of centimeters.
Image source: Network
It is understood that the platform mainly contains cameras, radar and lidar sensors three layers of positioning layer, camera positioning layer is the data directly collected by the vehicle camera, such as lane markings, traffic lights, road signs and poles, etc., the radar positioning layer is based on the radio wave backhaul of the aggregation point cloud to determine the precise position of the vehicle; lidar sensors can build a 3D world with a high resolution of 5 cm, combined with AI analysis radar data to locate surrounding objects.
However, the main reason why HD Maps can attract a number of giants to participate is not the "charm" of holographic intersections, but the strict demand for positioning accuracy of autonomous driving with a wider market scale further promotes its rapid maturity. Obviously, it is too early to talk about large-scale autonomous driving applications, but the application of scenarios such as unmanned distribution spawned by the epidemic has also become a testing ground for high-precision maps.
At present, the unmanned distribution field has gathered giants including SF, Meituan, Jingdong and so on, but unlike the Internet head enterprises have a large number of service businesses based on location information, Meituan, SF, and Jingdong have only begun to expand the unmanned distribution business after winning the high-precision map qualification in the past two years.
Write at the end
In recent years, with the continuous growth of the intelligent transportation market, coupled with the active promotion of local governments, upgrading and transformation projects have been released, and smart intersections and holographic intersections have also achieved a wider landing. At the same time, industry applications are also feeding back technology, and holographic intersections have undoubtedly opened up a new battlefield for high-precision maps in addition to autonomous driving.
Resources:
1. "The high-precision map in the giant photo, become the golden track", Internet of Things Media
2. "Good News | Huawei leads the release of the "holographic intersection" standard", Huawei machine vision
3. "Huawei Holographic Intersection, So That Every Vein of Traffic is Clearly Visible", smart transportation
4. "Smart/Holographic Intersection and Digital Twin", Smart Transportation
April 22, online
Amazon Cloud Technology AIoT Smart Home Event meets you