Transportation is the core link of human social operation, and driving is the most basic link in the field of transportation. For a long time, manual driving has been faced with problems such as fatigued driving, operational errors and illegal driving, which are also the main causes of traffic accidents. With the development of artificial intelligence, 5G, IoT, sensors and other technologies, autonomous driving technology has emerged as the times require, which is expected to completely solve the pain points of manual driving, restructure the traffic ecosystem, and stride towards the new era of unmanned driving.

Autonomous driving technology refers to the technology that realizes autonomous perception, autonomous decision-making and autonomous control of vehicles through on-board sensors, controllers, actuators and other devices, combined with artificial intelligence algorithms and communication technologies, and can complete driving tasks without human driver intervention. According to the classification of the Society of Automotive Engineers (SAE), autonomous driving technology is divided into six levels from L0 to L5, among which L0 has no autonomous driving function and is completely dependent on human driving; L1-L2 are assisted driving, which requires human drivers to monitor the whole process, and the vehicle can realize partial automatic control (such as adaptive cruise, lane keeping); L3-L4 are conditional autonomous driving, where the vehicle can realize autonomous driving in specific scenarios (such as highways, urban expressways), and human drivers do not need to monitor the whole process, but only need to take over the vehicle when necessary; L5 is full autonomous driving, where the vehicle can realize autonomous driving in all scenarios without human driver intervention.

At present, the development of autonomous driving technology has entered the critical breakthrough period of L2-L3 levels. Major global automobile enterprises and technology giants have successively deployed in the field of autonomous driving, continuously promoting technology research and development and scenario landing. From a technical perspective, autonomous driving vehicles mainly rely on the three core systems of “perception-decision-control” to achieve autonomous driving: the perception system collects real-time road condition information around the vehicle (such as vehicles, pedestrians, obstacles, traffic signals, etc.) through sensors such as cameras, lidar, millimeter-wave radar and ultrasonic radar, which is equivalent to the “eyes” of the vehicle; the decision-making system analyzes and processes the information collected by the perception system through artificial intelligence algorithms, and makes accurate driving decisions (such as acceleration, deceleration, turning, avoidance, etc.), which is equivalent to the “brain” of the vehicle; the control system converts the instructions of the decision-making system into the actual operation of the vehicle (such as controlling the accelerator, brake, steering wheel, etc.) through the actuator, which is equivalent to the “hands and feet” of the vehicle.

The application of autonomous driving technology will bring revolutionary changes to the field of transportation. First of all, it can greatly reduce traffic accidents and reduce casualties and property losses — according to statistics, more than 90% of global traffic accidents are caused by manual driving errors. Autonomous driving vehicles can avoid problems such as fatigued driving and operational errors, accurately identify road conditions, make quick responses, and effectively reduce the incidence of traffic accidents. Secondly, it can improve traffic efficiency and alleviate traffic congestion — autonomous driving vehicles can realize coordinated driving between vehicles, optimize vehicle speed and distance, reduce behaviors such as cutting in line and rear-end collision, and improve road utilization; at the same time, autonomous driving vehicles can realize automatic parking, reduce parking space occupation and improve parking efficiency. In addition, autonomous driving technology can also provide convenient travel methods for special groups such as the elderly and the disabled, improving the inclusiveness of travel.

However, the development of autonomous driving technology also faces many technical, ethical and legal challenges. Technically, the recognition accuracy of the perception system of autonomous driving vehicles in bad weather (such as heavy rain, heavy fog, heavy snow) and complex road conditions (such as construction sections, unmarked intersections) still needs to be improved; the artificial intelligence algorithm of the decision-making system still needs to be continuously optimized to ensure the accuracy and safety of driving decisions; in addition, the network security of the vehicle is also a major hidden danger. Once hacked, it may cause serious traffic accidents. Ethically, when facing the dilemma of “sacrificing the driver or the pedestrian”, the decision-making logic of autonomous driving vehicles has triggered widespread ethical controversies. Legally, a sound legal system for autonomous driving has not yet been established globally, and issues such as liability determination and insurance claims for traffic accidents still need to be clarified.

In the future, with the continuous breakthrough of artificial intelligence, 5G, sensors and other technologies, autonomous driving technology will gradually move towards the L4-L5 full autonomous driving stage. Countries around the world will strengthen technology research and development and scenario landing, improve the ethical supervision and legal system, and solve various challenges in the development of autonomous driving; autonomous driving vehicles will integrate deeply with intelligent transportation, IoT and other technologies, build an efficient, safe, convenient and low-carbon intelligent traffic ecosystem, completely change the way of human travel, and stride towards the new era of unmanned driving.