The source of driverless cars Automotive CAN bus simulation development tools

At present, the assisted driving functions that have appeared in cars generally use radar. This is relatively cheap, but the detection capability of radar is quite limited. The transmitting power of vehicle radar cannot detect very far. There is also fear of obstruction by objects, so driverless cars are used. How much practicality there is is also questionable. From manned driving to unmanned driving, it is divided into four stages.
The first stage is driver assistance. The driver assistance system can provide the driver with necessary information collection while driving, and give clear and precise warnings at critical times. Related technologies include: lane departure warning () , frontal collision warning () and blind spot warning system.
The second stage is semi-autonomous driving. When the driver still fails to take corresponding measures after receiving the warning, the semi-autonomous system allows the car to automatically respond accordingly. Related technologies include: emergency automatic braking (B), emergency lane assist.
The third stage is highly automated driving. This system can allow the car to provide long or short-term automatic control driving under driver supervision. This stage is still relatively preliminary.
Stage 4: Fully autonomous driving. The car can fully realize autonomous driving without the need for driver supervision, which means that the driver can engage in other activities in the car, such as working online, entertaining or resting.
According to this classification. The first stage has been basically popularized, and now it is basically available in high-configuration models, including BYD's 10,000-year-old model.
The second stage is becoming popular. European Union legislation requires that cars older than 3 months must be equipped with emergency automatic brakes (B). ’s city safety system, Honda’s B, and Mercedes-Benz are all at this level. Infiniti's new cars can also automatically control the steering wheel, bypassing the driver.
The third stage is currently in prototype. The new Mercedes-Benz series can have the computer follow the car itself in traffic jams, so the owner does not have to control the vehicle all the time. Some manufacturers, including China, are also doing relevant experiments and explorations.
The fourth stage is what Google has been working towards. The Google self-driving car like the light tank in the Red Alert game looks a bit weird. But after a long period of testing, it has reached a very high level of maturity and is the most practical self-driving car.
The principle of driverless driving and Google’s technical route
A driverless car can actually be regarded as a kind of robot. In principle, it is nothing more than sensors sensing road conditions and surrounding conditions, and then transmitting them to a computer to judge the situation based on artificial intelligence, and then notify the telex system. The telex system controls the mechanical device based on the signal, and finally the mechanical device controls the vehicle to perform various actions. .
In this process, the aspect of fly-by-wire mechanical control has been basically perfected, because in the past few years, the progress of the automotive industry has mainly been in this aspect. Nowadays, most cars are controlled by wire. Your brakes, accelerator, gear shift, and even direction are all a set of electronic signals. Your actions are interpreted and processed, and then transmitted to the mechanical system. In the past, what kind of direct mechanical to mechanical system The control can only be seen on very few tool trucks. Autonomous driving is nothing more than taking these signals away from human control and handing them over to computers for processing, which does not require much effort.
The technical difficulty lies in the first two steps. How to use sensors to accurately sense surrounding information? How does artificial intelligence make judgments?
From a technical perspective, there are many modes of sensing surrounding conditions. Google chose laser sensors. Laser sensors are very accurate in judging distance, but they are expensive and have limited practicality in bad weather. Currently, the cost of Google's self-driving cars is frighteningly high, mainly because of the laser sensors.