The world we see and touch is the real world,
The unseen and untouchable world is the virtual world.
On the first day of the China International Big Data Industry Expo held in Guiyang on May 26, the unique AI traffic commander of ZTE Softron Urban Traffic Brain attracted many guests to stop to visit and experience. What are the innovative ideas and dark technology applications of AI traffic commander? The following Xiaobian take you to find out.
A traffic commander running around in the real world
We all live in the real world of people, cars, roads, devices, etc., but we also live in the virtual world surrounded by numbers. We hope to have A vehicle that can travel between the virtual and the real like Doraemon's time machine to explore the past and the future. Zte Softran AI Traffic Commander is playing such a role, by connecting the virtual world and the real world, so that we can freely shuttle between the time and space of traffic.
Zte Soft Chuang AI Traffic Commander is the right-hand man of urban traffic managers. As an intermediate carrier connecting virtual and real urban traffic brain, AI commander fully integrates the super perception ability and innovative thinking ability of traffic brain to create a control center with self-evolution ability, realizing intelligent matching of real traffic world plan, automatic generation of instructions, intelligent scheduling of traffic elements and other functions.
Through the simulation of massive traffic models, the intelligent dispatching function of AI traffic commander constantly learns and analyzes the behavioral characteristics, psychological characteristics and other indicators of traffic participants (government, enterprises and people), seeks the game balance point among traffic participants, and thus reasonably schedules traffic resources and guides the travel behaviors of traffic travelers. To realize the real-time, accurate and efficient all-round autonomous regulation and development of the entire transportation ecosystem.
Reconfigured the AI traffic commander's control strategy for tough traffic battles
Traffic command is a process of "knowledge" + "experience" + "principle". The so-called knowledge is the objective law and the objective relation existing in traffic management. And experience, is in the long-term actual combat work, countless mechanical and repetitive work summed up in the natural reflection; Principles are established rules to be followed when dealing with sudden or new phenomena.
AI traffic commanders should first systematically integrate effective theoretical knowledge of traffic engineering and transportation engineering, fully organize, learn and accumulate repetitive work and immediate measures in daily traffic management, and finally formulate management principles in line with regional status according to local conditions. Then, they should improve the existing command, control, monitoring and early warning, and efficiency improvement step by step.
Furthermore, it is necessary to integrate data, sort out effective information, and reduce repetitive work in daily traffic management, such as frequent rotation of the camera to find and focus on the target, repeated modification of signal schemes to deal with events, repeated submission of multiple information sources for the same event, and troublesome cross-service scheduling actions. Relying on the data analysis and integration capabilities of the big data analysis platform, ZTE Soft Creation provides auxiliary modules such as camera auto focus, automatic execution and cancellation of signal induction schemes, dispatch relay of multi-service platforms, etc., which can be implanted into the existing traffic system quickly and imsensibly, retain the existing usage habits, but improve the accuracy of data, reduce repetitive labor, and provide convenience.
At the same time, traffic management is a very rigorous and pragmatic management technology. Because of this, there are still a large number of police officers, technical managers and front-line command staff doing similar work day and night at the present stage. It is necessary to use AI traffic commanders to reduce manual intervention and improve the level of automation. The automatic release of induction, automatic adjustment and optimization of signals, automatic allocation of facilities, automatic scheduling and deployment of police can achieve automatic operation for 80% of the time, greatly reducing the working pressure of front-line personnel, and improve the level of technology and response speed of urban traffic management.
Remarkable ability, AI traffic commander's three tricks
In order to deal with the problems of normal urban traffic congestion and frequent traffic incidents, we need AI commanders to have the following capabilities to ensure and improve the overall traffic operation safety:
One-button command
The road perception information obtained from the perception center and the socialized traffic travel data from the Internet are applied to the dual drive of the computing engine and algorithm engine of the thinking center to realize the rapid perception of traffic accidents and congestion conditions and the urban emergencies in an instant and comprehensive way, which can realize the second-level alarm of blocking points and event discovery, replace human patrol and find traffic anomalies. The AI commander can make a quick judgment on the situation and progress, and at the same time, combined with the intelligent drive tool, issue the work order to deal with the situation, timely transmit to the police officers in the duty area, and select the appropriate police officers to deal with the police situation. At the same time, AI commanders track the processing process and time of events to ensure the operation results of urban traffic.
Independent regulation
AI commanders can deduce and simulate evaluation of the event handling process according to the operation condition of the road network and the progress of event handling, combined with high-precision maps and simulation tools. AI commanders can realize the overall control and management of road traffic flow, speed and saturation degree according to the simulation evaluation results, and real-time optimization of signal timing scheme. In addition, relevant guidance information release strategies are formulated, and the policy suggestions are studied and demonstrated for many times to guide the traffic balance of the road network.
Continuous evolution
AI commander has autonomous learning function, which can continuously train and optimize the AI commander according to the evaluation of historical event disposal. It can break through the calculation ability and basis of the thinking center, carry out routine and convenient disposal of events and situations, enhance his memory function, realize the quick call of copies and the fine management of scheduling.
In the real world, AI traffic commanders can run much more accurately
In the real world, AI commanders gradually adapt to the rapid development of traffic and take it as their duty to improve the efficiency of traffic management and reduce ineffective running of police officers. So far, AI commanders have been successfully implemented in Ningbo, Nanjing, Shijiazhuang and other cities. Among them, Ningbo Traffic Police intelligent traffic business support platform and Zhongxing Soft Chuang AI Traffic Commander Control center combined to achieve the average number of waiting red lights at key intersections in the urban area from 2.5 to 1.8 times, congestion ranking from 44 to 70, congestion relief time reduced from 15 minutes to 11 minutes. To achieve the rapid identification of traffic signal problems, timely correction, rapid tuning, as well as the optimization and accurate implementation of traffic organization.
