Driverless cars are expected to be on the road in more than ten years. The bottleneck of technology and cost needs to be broken through.
driving is sometimes boring. During the peak period of congestion, people hope to get rid of frequent braking and starting and find the fastest way home; In the tired long-distance driving, people hope they can doze off and wake up safely
let the car drive by itself is desirable, but the unmanned car is still considered too far from the market. However, in October 2010, Google, a network giant with strong research and development resources, revealed that its self driving prototype had been driving 140000 miles quietly. This move is puzzling, but the company's past successful investment in Android, youtube and doubleclick shows that the company has the ability to turn its non search business into a new major source of revenue. This has drawn much attention to the future of driverless technology
driverless cars run well
in fact, driverless technology has been studied abroad since the 1980s. For example, the navlab series prototype car developed by Carnegie Mellon University in the United States covers a variety of models, such as minibuses, trains, multipurpose vehicles, large buses and so on. In 1995, the navlab-5 prototype car crossed the United States, with a total journey of 4587 kilometers, of which autonomous driving accounted for 98.2%. The longest continuous autonomous driving distance was 111 kilometers, and the average speed of the whole journey was 102 kilometers per hour. For another example, the unmanned vehicle cooperated by the National Defense University of Munich and Mercedes Benz also carried out a 1600 km long-distance experiment, 95% of which was self driving and performed more than 400 lane changing operations
in 2007, the defense Advance Research Projects Agency (dapar) held a competition called the grand challenge, which is a milestone in the history of driverless technology. In the environment of completely imitating urban roads, 11 teams stood out and participated in the final. The organizer arranged 30 manned cars in the stadium to imitate the actual road conditions. The total length of the race is 96 kilometers and the time limit is 6 hours. Finally, Stanford University was the first to cross the finish line, followed by Carnegie Mellon University. However, the award of the competition is not based solely on driving time, but on comprehensive consideration of safety and stability. Finally, Carnegie Mellon University and Stanford University ranked first and second respectively. A total of six teams completed the whole journey, and the results were very encouraging
it is reported that Google's unmanned vehicle development plan is to absorb researchers from Stanford University and Carnegie Mellon University
in July 2010, the unmanned vehicle developed by the University of Parma, Italy, traveled 13000 kilometers across two continents and arrived in Shanghai at the end of October to participate in the World Expo. However, this experiment is actually a combination of two cars. The first car is driven by people and reports the positioning to the second car through the GPS navigation system
in China, many units such as the National University of defense technology, Nanjing University of technology, Zhejiang University and Tsinghua University also began to track and study driverless technology very early. Among them, the smart car thmr-v developed by Tsinghua University has an average speed of 100 kilometers per hour and a maximum speed of 150 kilometers per hour when the lane line is automatically tracked. FAW and the National University of Defense Science and technology jointly developed the Hongqi car as the prototype. Under normal traffic conditions, the maximum stable speed of the Hongqi car running on the highway is 130 kilometers per hour, and the maximum peak autonomous driving speed is 170 kilometers per hour, with overtaking function. Moreover, the control components are miniaturized and integrated with the original vehicle system, maintaining the original style of the carriage
in 2008, the National Natural Science Foundation of China set up a major research program related to intelligent vehicle technology, cognitive computing of visual and auditory information. In 2009 and 2010, the Foundation Committee initiated and hosted two Chinese smart car future challenge competitions. In each competition, more than ten units, including wood chip raw materials, pulp and finished paper, sent people and cars to participate in the current round of tariffs imposed on the United States, resulting in fierce competition
technology and cost bottlenecks need to be broken through
however, although many unmanned vehicles have achieved good results, they are still far from enough in terms of adaptability to complex environments, reliability of operation and cost control, which are exactly the important problems that unmanned vehicles must face when they are officially introduced to the market. Before these problems are fundamentally solved, neither manufacturers nor customers can dispel relevant concerns
first of all, the most important thing for unmanned vehicles on the road is reliability and safety, especially the direction, throttle, brake and other control systems and cameras Lidar etc "The development of PU materials is to create a wide range of attribute measurement components. As we all know, the working temperature requirements of cars are very demanding, from tens of degrees below zero to tens of degrees above zero, while the working temperature of commonly used sensors such as cameras and industrial computers is generally between zero and 50 degrees, which is far from meeting the reliability requirements, and there is an urgent need to adopt new technologies, new methods and new materials.
the development of unmanned vehicles should be coordinated with the automotive industry as soon as possible Cooperate or combine to design, debug and install according to the technical standards of vehicles, so as to improve the reliability of new unmanned vehicle equipment and devices
in addition, the performance and technology of various sensors and computers used in unmanned vehicles have made great progress compared with 10 years ago, and the price has also been significantly reduced, especially the price reduction of computers is obvious and prominent. Nevertheless, the current price of unmanned vehicle equipment or devices is still high, which is difficult for the public to bear. It costs tens of thousands to hundreds of thousands of yuan to buy an infrared camera or lidar, while it only costs tens of thousands of yuan to buy a low-end car
especially for China, almost all the cameras, radars and other products used by unmanned vehicles now rely on imports. The state should increase investment, strengthen basic research, strengthen raw material research, and establish a relatively complete production base of sensors for unmanned vehicles
in terms of the current technical level, it should be a more appropriate way to gradually transform the scientific and technological achievements formed in the development of unmanned vehicles into partial improvements to existing vehicles. In fact, the transformation from manual gear to automatic gear, reversing radar, road detection based on vehicle camera and other vehicle assisted driving technologies have been successfully applied in commercial vehicles. The popularization and application of auxiliary driving system is conducive to increasing the output and reducing the cost of unmanned vehicle parts. And with the gradual application of these technologies, driverless driving will become a reality imperceptibly, and people will feel that driverless cars are a normal thing
if we estimate optimistically, we may see that unmanned cars are really on the road in more than a decade
legal ethics: establishing standards is the key
in addition to technical reasons, the legal ethics of unmanned vehicles also poses new challenges. Unmanned vehicle belongs to the category of intelligent robot in essence, and should abide by the famous three principles of robot:
first, robots should not harm humans, and robots should not be ignored to harm humans
second, robots should obey human commands, except for those contrary to the first one; Third, robots should be able to protect themselves, except for those in conflict with Article 2
although it is a little premature to consider these factors at present, after all, the unmanned vehicle is far from reaching the stage of competing for control with humans, in this technology 7. The transmission chain of the compression test machine should be filled in time, and the lubricating oil should be No. 50 mechanical oil. When filling, the test machine can be tilted backward to make its base tilt up, and the brush with a brush dipped in oil on the chain at the lower part of the base. The oil should not be excessive, so as not to flow downward In the process of gradual maturity, commercialization and marketization of the technology, if there is a lack of clear definition of the law, it is a major constraint on the introduction of unmanned vehicles to the market
as for how to define the law, it is still a controversial issue. However, in accordance with the law of any product going to the market, a basic premise is that in order to go to the market, relevant international and domestic standards must be established, and only those that meet the standards can go on the road
at present, countries all over the world are stepping up the development of unmanned vehicles, which is very beneficial to the formation of standards; At present, China is also catching up, which is also very beneficial for China to establish a position in the process of formulating standards. However, unlike foreign countries, domestic research institutions that develop unmanned vehicles generally lack contact with automobile manufacturing enterprises, or automobile manufacturing enterprises have little interest in driverless driving. This problem deserves attention to these markets. Because the formulation of standards for unmanned vehicles cannot be separated from automobile manufacturing enterprises, and once the technology is mature, unmanned vehicles are also related to the survival of automobile manufacturing enterprises
in addition, due to the fact that driverless state and manned state often alternate during driving, once an accident occurs, it will be very difficult to define it. This may require the installation of powerful black boxes in unmanned vehicles to be defined by data analysis after an accident. If it is the operation error of the personnel in the vehicle, it must be attributed to the operator; However, if the accident does occur in the process of driverless driving, it can only be borne by the manufacturer of driverless vehicles
LINK
Copyright © 2011 JIN SHI