The emergence of advanced driver assistance systems has made autonomous driving closer and closer to us. With the maturity of autonomous driving technology and the acceleration of commercialization, the car is not only a means of transportation, but gradually transforms into a diversified role of intelligent entertainment hardware. When consumers buy a vehicle, safety is not only the only indicator, but also comfort. Sex and entertainment have also become considerations when buying a vehicle. In the future, the emergence of fully autonomous driving will promote another revolution in the automotive consumer market. Self-driving cars will change people’s travel habits.
At this stage, autonomous driving technology is not yet mature, and there are still many possibilities for the development of autonomous driving. Among them, bicycle intelligence and intelligent network connection are the two major technical routes for the development of autonomous driving technology.
Bicycle intelligence mainly relies on the vehicle’s sensors such as millimeter-wave radar, lidar, and vehicle vision cameras, wire control systems, and computing unit hardware for environmental perception, decision-making, control and execution, allowing vehicles to realize independent thinking and decision-making like humans. Drive the vehicle to the intended destination.
Intelligent network connection is to link the elements of “people-vehicle-road-cloud” together through the Internet of Vehicles technology, so that the functions of environmental perception, decision-making, control and execution of the vehicle can be upgraded, and the automatic driving function can be improved and promoted. The management and improvement of the transportation network can provide safer, more comfortable, more energy-efficient, and more environmentally friendly driving methods, promote the adjustment of the urban transportation system, and build a new type of smart city.
The development model of bicycle intelligence mainly relies on the coordination of hardware equipment and software installed on the vehicle to realize the automatic driving function of the vehicle. This development model of bicycle intelligence has higher requirements for the perception of the environment, in order to make bicycle intelligence safer , It is necessary to install enough hardware equipment on the vehicle to eliminate the visual blind zone during the driving of the intelligent bicycle. However, the development of bicycle intelligence has not been able to achieve large-scale commercial use. The main reason is that the bicycle intelligence is very intelligent. It is difficult to achieve sufficiently high security.
In order to obtain a better perception effect, it is necessary to install on-board vision cameras, ultrasonic radars, millimeter wave radars and other hardware devices in the vehicle. However, in the actual use process, there will still be blind spots, which cannot be recognized in all directions, and the road conditions are complicated. Roadside obstacles cannot be identified by these hardware. If there is an object behind the obstacle to make the obstacle move, and the vehicle does not make a pre-judgment, it will lead to more dangers. In addition, the requirements for the installation location, field of view, data throughput, calibration accuracy, and time synchronization of the perception hardware equipment are also very high. Intelligent bicycles drive in busy intersections, bad weather, small object perception and recognition, signal light recognition, backlighting and other environmental conditions. However, it is still difficult to completely solve the problems of accurate perception and recognition and high-precision positioning.
In addition, the current vehicle-mounted visual cameras, ultrasonic radar, and millimeter-wave radar have a very limited sensing range, and the safety distance of the vehicle needs to be detailed to the centimeter level or even the millimeter level. If there are vehicles on the road very fast, the vehicle does not do it in time. To predict, it will cause the vehicle to fail to respond in time, leading to the occurrence of danger. Many major car factories now provide L2 and L3 level automatic driving. When encountering this situation, the driver needs to control the vehicle in time. If encountering similar problems, the driver will not be able to make a timely takeover response.
The development of bicycle intelligence is inseparable from high-precision maps. The collection of high-precision maps is very complicated and very costly. If the road is rectified or repaired, the high-precision maps need to be updated in time. If the routes of bicycle intelligent autonomous driving can be commercialized The high-precision map needs to be updated at any time according to the road conditions. This update frequency requirement is even higher, which cannot be achieved at this stage.
Excessive cost is also one of the main reasons hindering the intelligent development of bicycles. Because the cost of hardware devices is too high, and in order to process the signals generated by these hardware devices, computing units need to be configured, and corresponding software systems and software systems will be deployed on the car side. The maintenance management and update of the auto-driving vehicle also greatly increase the maintenance cost of the self-driving vehicle. At present, the hardware equipment of L4 autonomous vehicles generally includes: 6-12 vehicle-mounted vision cameras, 3-12 millimeter wave radars, less than 5 lidars, 1-2 GNSS/IMUs, and 1-2 computing platforms. These combined costs are very high, which makes the intelligent commercialization of bicycles more difficult.
There are now two feasible solutions for bicycle intelligence,
One solution is represented by Google, which is equipped with Lidar, millimeter-wave radar, vehicle vision camera and other perception hardware in the vehicle. The disadvantage of this solution is that the cost is too high and the possibility of commercialization is low, but multiple perceptions The coordination of hardware has also increased the safety of autonomous driving, which has been sought after by many major car manufacturers. It is also the main approved solution for many OEMs at this stage.
The other is a solution represented by Tesla that realizes autonomous driving through the collaboration of on-board vision cameras and software. This solution is relatively low in cost and has a high possibility of commercialization. It has been seen at this stage. The prototype of commercialization, but only based on the on-board vision camera to achieve automatic driving, less information sources are obtained, and safety cannot be guaranteed.
Although the technologies used in the two schemes are very different, they are all based on the smart bicycle route.
As the development of bicycle intelligence technology enters the bottleneck, intelligent network connection is gradually mentioned. Intelligent network connection refers to the organic combination of the Internet of Vehicles and bicycle intelligence. Information exchange and sharing between vehicles and people, vehicles and roads, vehicles and back-end (V2X), etc., to realize information communication between vehicle driving information and the main vehicle factory, management department, and destination facilities, and promote the realization of safety, comfort, energy saving, Efficient driving solutions will finally realize the commercialization of fully automated driving.
The technical solution of intelligent network connection is to make vehicles become smarter to become smarter for cars and road facilities, from bicycle intelligence to group intelligence. Due to the high cost of commercialization of bicycle intelligence, roadside equipment is used to replace some of the technologies to make road facilities. Being smart can effectively reduce the cost of vehicle manufacturing and R&D and realize the possibility of commercialization as soon as possible. The prerequisite for intelligent network connection is the intelligent transformation of road facilities and infrastructure investment, which requires participation in many aspects, such as whether the government management department will support the intelligent transformation of road facilities, and whether the main vehicle factories are willing to The sharing of vehicle driving information, whether investors will support such technical solutions and other factors will affect the development and evolution of intelligent network technology.
The development of intelligent network connection is not possible overnight. It needs to be realized step by step. It is mainly divided into three stages. The first stage is the information interaction and coordination stage, which can realize the information interaction and sharing between vehicles and roads, such as collision warning and road hazard warning. The second stage is the cooperative perception stage. On the basis of the first stage, the vehicle and road facilities realize the perception and positioning, and the third stage is the cooperative decision-making stage. Control, so as to realize the possibility of fully automatic driving. The realization of intelligent network connection allows vehicles to collect road information in advance during driving, so as to predict driving behavior.
my country’s intelligent network connection research and development is rapid. In the first stage, the intelligent network connection has carried out large-scale test verification and pilot demonstration in many cities, and gradually carried out the first trial of commercial operation; the second stage is the construction in some cities In deployment and testing, the road’s collaborative perception capabilities have been fully verified, and the coordination and control of infrastructure in some scenarios have also been explored. However, the current technological development is still unable to realize the commercialization of fully automated driving.
Bicycle intelligence and intelligent network connection are both effective autonomous driving solutions. On the whole, autonomous driving under intelligent network connection is more likely to be achieved, but intelligent network connection touches many areas and requires coordination. There are also complicated roles. It needs the coordination of various departments and enterprises to realize and landing.