Tencent Shi Xuesong: Create a two-wheeled digital base driven by AI to help transform the transportation industry

On September 8, the 2023 Tencent Global Digital Ecosystem Conference Smart Transportation Special Session was held in Shenzhen. Shi Xuesong, Vice President of Tencent Intelligent Transportation, delivered a keynote speech titled "Tencent Digital Base, Helping the Transformation of the Transportation Industry."

Currently, transportation development is shifting from scale growth to high-quality growth, and transportation technology innovation has become the main driving force. As the level of transportation digitalization continues to improve, the industry is paying more and more attention to the in-depth application of digital technology, focusing on intelligence to improve infrastructure efficiency, transportation safety and performance, and reduce costs.

Tencent Intelligent Transportation Vice President Shi Xuesong is the person in charge of Tencent Intelligent Transportation

Shi Xuesong pointed out that in the future, transportation life forms need to build a closed loop from full perception to calculation, decision-making, and access, and have the ability to evolve on their own. In this regard, Tencent proposed the AI-driven two-wheel architecture as a universal technical base for digital transportation.

Shi Xuesong said that real-time twins and traffic OS are two key components of the smart transportation design architecture. Real-time twins provide high-quality and high-value data foundation for large artificial intelligence models by recording dynamic scene data of transportation. Tencent also makes full use of existing digital tools to break down digital islands, host applications in the form of traffic OS and conduct automated management. Based on this foundation, the design architecture of smart transportation can be reconstructed

Shi Xuesong emphasized that the construction and iteration cycles of transportation infrastructure and digital facilities are not synchronized, so in the process of digital transformation, it is necessary to break information islands with a layered and decoupled design to achieve rapid evolution at all levels. Tencent hopes to make full use of the technology and capabilities accumulated over the past two decades, adhere to the "people-oriented" concept to develop smart transportation, implement product methodologies, and bring safer, more efficient, and better travel services to the people.

The following is the full text of Shi Xuesong’s speech:

Good morning everyone, I am Shi Xuesong from Tencent Smart Transportation.

China’s construction of a transportation power has entered a new stage targeting high-quality development. High-quality development includes high-quality operations and high-quality services for travelers. Today I will share some experiences based on Tencent’s insights and practices in smart transportation construction

In April this year, the Ministry of Transport, the State Railway Administration, the Civil Aviation Administration of China, the China Post Bureau and the China Railway Group jointly released the "Five-Year Action Plan to Accelerate the Construction of a Powerful Transportation Nation", marking the entry of digital technology into the construction of a powerful transportation country. A new stage of integrated development with the transportation industry. Tencent proposed to build a future transportation life form in 2020. In 2021, we proposed to use a digital base to achieve this goal, and adopted an AI two-wheel drive strategy to build it. One wheel of the two-wheel drive is the real-time twin, and the other is the traffic operating system.

Tencent’s real-time twin combines and records dynamic and static traffic scenes through digital twins. With the help of the real-time twin method, we can accumulate a large amount of valuable data in the transportation system, which will become an important basis for processing large-scale models. In the transportation industry, real-time twins can create valuable and high-quality data for us. We can also make full use of existing digital tools to break down information silos and present them as a transportation operating system. When we have a large amount of data and establish an application environment, we can use artificial intelligence tools to realize a closed loop from full perception to calculation, decision-making, and access, and become a future transportation life form that can evolve itself

Comprehensive perception is very important in transportation. When providing transportation services, we need to have a good awareness of transportation objects and operating entities. Because transportation not only involves complex infrastructure, but more importantly, it carries mobility, including logistics, people and vehicles. If we could describe these perceptual results in a semantic way, that would be the best language and it is very critical for us. A typical application of perception is in the field of containers. Through visual processing and analysis, we can realize the application of intelligent container inspection

Calculation is very important. The core symbol of Tencent's digital base is computability. For example, in the direction of transportation, how to integrate and interact with Internet big data and real-time data generated during transportation operations is a direction Tencent has been exploring.

Today, in this direction, we have gradually moved towards a closed loop. For example, we built a base using BIM GIS, which solves spatial calculation problems. However, in Tencent's real-time twin system, BIM GIS is only static, and we also need to take time into consideration to make it a spatio-temporal calculation

After adding the time dimension, traffic scenes can be depicted and recorded. Otherwise, only information can be recorded, but the scene cannot be recorded. Therefore, through the combination of BIM GIS, spatial calculation is realized, and after adding perception and time relationships, it is upgraded to four-dimensional spatio-temporal calculation. After completing the construction of this digital base, we can generate a large amount of high-quality data, including transportation scenarios

In terms of traffic control and guidance decision-making, once the digital processing is completed, we can use these high-quality data to conduct decision-making deductions. In this regard, we use Tencent's simulation technology and big data analysis technology, and with the help of Tencent's Traffic OS, we can also carry out automated management and control. Using these data for traffic simulation and analysis, the convergence speed is very fast and the improvement is huge. In this direction, we use the real-time twin method to restore the dynamic and static aspects of traffic, so that many new scenarios can be created, which, combined with existing real-time data, can have a good impact on traffic control. This method can also be extended to more scenarios, combined with more models, to derive better methods and means. Tencent also provides Traffic OS, which is both a tool and an environment. Through low-code methods, traffic control applications can be created through drag-and-drop operations. In this way, ordinary operators can build transportation applications across code barriers

Touch is the most critical thing today, it is how we provide high-quality services for travelers. Tencent's philosophy is people-oriented, which means that every user who uses Tencent's travel services on the road must feel that the services we provide are good.

The automotive industry now regards cars as very important products, attaching great importance to the interaction with manufacturers and car owners as well as the construction of touch points in the car; similarly, our roads are also a product, we are not just building It is very important to build a long-term communication relationship with all users who use this path

For example, if the road is expanded to a two-way 16-lane road, then how to better guide vehicles becomes very important. The current navigation system may say "Turn right 250 meters ahead and don't go uphill." However, it is difficult for many people to accurately understand the distance of 250 meters and it is easy to miss the exit. But if the navigation is simplified to the three simple instructions of "change lane right, change lane left, execute", it can be guided more accurately. However, this requires the navigation system to know exactly where the vehicle is. We use real-time twin technology to provide accurate location and situational awareness for each moving vehicle. With just an ordinary communication terminal in the vehicle, we can clearly demonstrate the relationship between people, vehicles and roads. In this way, even complex bootstrapping can be simplified to implement

When real-time twin technology reaches a certain level, we have a good platform architecture and tools. It is very important for road operators to allow command and monitoring screens, field operators and end users to obtain data from the same source. If a traffic accident occurs in one of the 16 two-way lanes and needs to be closed and guided, rescue vehicles can use this method. Therefore, through the real-time twin platform, road operating companies have been able to precisely interact service capabilities with specific users, which is an extremely valuable resource. At this time, we can use product methodology to build new designs

After building these bases, we can continue to accumulate various scenarios, various management and control capabilities, and various service capabilities on them, and can continuously and automatically evolve forward. This is a very important capability. After building an information platform, it used to be maintained by IT engineers; today's base is different from the past. In addition to routine and typical operation and maintenance work, more importantly, it can evolve on its own.

In the development process of smart transportation, we must be aware of an important issue, that is, the construction and update cycle of transportation infrastructure far exceeds the cycle of information systems and digital systems, and they do not match each other. Therefore, during the construction process, we must carry out hierarchical decoupling and achieve leap-forward sharing

Rewritten content: Today, Tencent uses distributed cloud technology at the bottom layer, uses cloud-native digital twins and transportation operating systems at the upper layer, and then deploys artificial intelligence platforms to achieve good layering and support different levels. rapid evolution. It can also continuously improve its capabilities through its own accumulation. Many problems can be solved through a layered approach. Some people believe that large systems should also integrate software and hardware, but this can easily lead to isolation. During the digital transformation of the transportation industry, the integration of software and hardware in such a large system is prone to problems. Therefore, in the direction of transportation intelligence, if we can establish a solid foundation, adhere to the people-oriented approach, and implement the product methodology, we can achieve good results

The above is what I have shared with you based on my experience in the past few years. Thank you all.

The above is the detailed content of Tencent Shi Xuesong: Create a two-wheeled digital base driven by AI to help transform the transportation industry. For more information, please follow other related articles on the PHP Chinese website!