Recently, three patents about Google's self-driving car were exposed. One is to switch the unmanned state, one is to remind passengers, and the other is to gesture the car. Google's driverless car is getting closer and closer to us. .
Since Google started the driverless car project, the industry is very interested. Originally, the driverless car should be the domain of the car giants. The Internet company Google is going to have a foothold. The patent exposure seems to be not far from the success of the project. How much value does the second patent have? What will the future driverless car look like? Let's take a look.
Google Autopilot Test Vehicle
First, the source of driverless cars
The so-called driverless car is not falling from the sky, but with the development of the automobile industry, the advancement of electronic control technology has gradually evolved from manned assistance.
From manned to unmanned, the industry is divided into four phases.
The first stage is driver assistance. The driver assistance system provides the necessary information for the driver to drive while driving. At the critical time, clear and precise warnings are given. The related technologies are: lane departure warning (LDW), frontal collision. Warning (FCW) and blind spot alarm system.
The second stage is semi-automatic driving. When the driver still fails to make corresponding measures after receiving the warning, the semi-automatic system can automatically respond in the car. Related technologies include: Emergency Automatic Brake (AEB), Emergency Lane Assistance (ELA).
The third stage is highly automated driving. The system can provide the car with long-term or short-time automatic control driving under the supervision of the driver. This stage is still relatively primitive.
Stage 4: Fully automatic driving, the car can fully autopilot without driver supervision, meaning that the driver can engage in other activities in the car, such as online work, entertainment or rest.
Follow this classification. The first phase has been basically popularized, and now the high-profile models are basically brought, including BYD in the early 100,000.
The second phase is gaining popularity, and EU legislation requires cars to be fitted with emergency automatic brakes (AEB) before November 2013. Volvo's urban security system, Honda's CMBS, and Mercedes-Benz's Pre-Safe are all at this level. Infiniti's new car has also been able to bypass the driver's automatic control of the steering wheel.
The third stage has a prototype, the new Mercedes-Benz S-system can be in the case of traffic jams, the computer itself follows the car, the owner can not always control the vehicle. Some manufacturers, including China, are also doing related experiments and explorations.
The fourth stage is the goal that Google has been working hard. The Google driverless car in the Red Alert 2 game is a bit strange. However, it has been tested for a long time, and its maturity is already very high. It is the most practical and unmanned car.
Second, the principle of driverless and Google's technical route
A driverless car can actually be considered a robot. In principle, the sensor only senses the road condition and surrounding conditions, and then transmits it to the CPU. The CPU judges the situation according to the artificial intelligence, and then informs the telex system. The telex system controls the mechanical device according to the signal, and finally the mechanical device controls the vehicle to do each. Kind of action.
In this process, the control of telex machinery has basically been improved, because the advancement of the automotive industry in the past 20 years is mainly in this respect. Most of today's cars are controlled by telex. Your brakes, throttles, shifts, and even directions are all a set of electronic signals. The actions you operate are interpreted and transmitted to the mechanical system. In the past, which direct mechanical to mechanical The control is only visible on very few tool carts. Driverlessness is nothing more than a matter of taking the signals out of the human control and handing them over to the computer, without too much trouble.
The technical difficulty lies in the first two steps. How can the sensor accurately sense the surrounding information? How does artificial intelligence make judgments?
From the technical route, there are many modes of sensing the surrounding situation. Google chose the laser sensor. The laser sensor judges the distance very accurately, but it is expensive, and the utility is limited in bad weather. At present, the cost of Google's driverless cars is high and scary, mainly because of laser sensors.
The cheaper one is the optical camera. At present, the Mercedes-Benz S-class magic carpet technology uses the camera as the information acquisition source to control the air suspension. However, if the optical camera is used for driverless, the artificial intelligence for image recognition will be very high. The judgment of speed will also be very troublesome.
At present, the auxiliary driving function that has appeared on the car, the radar generally used, this is cheaper, but the radar's detection capability is quite limited, the launch power of the car radar can not detect how far, and the object is blocked, and the driverless car is used. How practical is there is also doubtful.
From the current point of view, the most reliable is Google's laser sensor route.
The degree of artificial intelligence is even higher. In fact, driverless driving on the highway is not too difficult. Google has no problems now. Mercedes-Benz and Audi have no problems. VOLVO has no problem. Even the unmanned high-speed driving in China has run 286 kilometers without problems. This means that the situation of the highway is simple and the algorithm required for artificial intelligence is not very demanding.
Driving in the city is completely different.
Google’s driverless car project director, Emerson, said, “It’s much more complicated to travel a mile in an urban area than on a highway because the former has hundreds of different rules in a small area. The road surface situation. We spent a lot of time improving our software, so it can now distinguish hundreds of different objects in real time - such as pedestrians, buses, parking signs by coordinators, or cyclists Give the gesture signal to pass the road, etc."
In terms of artificial intelligence, other vendors are still far from Google, Google's artificial intelligence level is close to practical, and other manufacturers are still in the performance stage.
Third, the value of exposure patents
One of the three patents exposed this time was a patent called "Determining when to drive autonomously" that Google applied in March 2014. This patent is actually a continuation of a patent in 2012. The specific content is to collect information through various sensors mounted on the car, and compare the map data with the traffic model. Once there is a large deviation from the car, it will automatically report to the driver. A warning is issued to remind the driver to perform human control. If the passenger is uninterrupted, the self-driving car will automatically reduce the speed or enter a safer lane.
This patent is regarded as a safety patent, requiring manual intervention. If the labor is not involved, then reduce the speed into a relatively safe lane to avoid accidents.
From the level of artificial intelligence, it is a matter of several lines of code, not high technology, this patent is not a core patent.
The second is a patent that Google applied for in 2013. When the car detects a pedestrian passing in front of it, it will respond correctly, such as slowing down or even stopping, and then issuing a car to the pedestrian. The signal to be made, the signal may be a light, an electronic signage or a sound.
The purpose of this patent is to allow communication between the car and the pedestrian to solve the problem that the driverless car is difficult to judge the intention. From the perspective of artificial intelligence, it is only detecting the pedestrian and then signaling, and there is not much value.
The third is automatic gesture control. Google has applied for a patent called “3D Gesture Automatic Controlâ€. A 3D camera is installed at the top left of the driver's seat. It can monitor and recognize some gestures of the driver, thus activating some functions. . According to the patent description, gestures can be used to adjust the seat, change the fan speed, adjust the air conditioning temperature, switch windows, and more. For safety reasons, some functions involving driving operations require physical button operations.
In addition, the driver can learn preset gestures. Since the system supports face recognition, different drivers can also set gestures for individuals.
This thing is almost a replica of the face recognition on the mobile phone and the somatosensory control in the game machine. It is technically already commercialized, but Google considers the convenience of the driverless car and applies it to the car to apply for a patent. . Limited practicality, not as good as voice manipulation and recognition. Not even a core thing.
Therefore, this time Google’s three patents were exposed. If it is said that the technical information is leaked, it is better to say that it is an advertisement. The real core of Google's driverlessness is the algorithms for sensor acquisition processing and artificial intelligence. The three patents exposed this time are not of much value.
Fourth, the future of the car
Through the above analysis, we can see. On the road from assisted driving to unmanned driving, Google is currently at the forefront.
While other vendors are still exploring the third phase of highly automated driving, Google has been working in the fourth phase for several years. When other manufacturers can only perform on the highway, Google has solved the city road driving with a hundred times more difficulty.
Through the study of the principle of unmanned driving, we found that Google's current exploration is not limited to Google cars, and can be transplanted to any new car.
As long as the car is controlled by telex, you can transplant Google's signal acquisition and artificial intelligence. It is only the power, braking and direction parameters of different cars.
Whether it is an electric car, a fuel car or a hybrid car, it is possible to transplant Google's driverless system without any difficulty.
When Google succeeds, BMW, Mercedes-Benz, Audi, Tesla, Bentley, Toyota, Honda, Nissan, Mazda, Great Wall, Changan, BYD, Chery, no matter which car manufacturer, just use Google's solution to buy Google's algorithm, Buying the hardware provided by Google can achieve driverless functionality.
In the future, the status of driverlessness will be similar to today's automatic car and automatic parking. As long as you are willing to pay, any model with a model of a model with a high version can be provided. Driverless is not a new car, but can be installed on any new model.
When unmanned driving is popular, some new functions and industries will be derived. Because the popularity of unmanned driving means that each car has a high-performance car, through the future 5G network, they can communicate with each other, can communicate with the city management center, and even communicate with the sensors on the road. People can manage cars like high-speed rail, and traffic accidents become a historical term.
When the communication between the car and the restaurant, hotel, gas station and charging pile is established, the car can automatically refuel and recharge, and bring the owner to the restaurant and hotel.
When the car is connected with the mobile phone, tablet, and computer, it can issue instructions to the car at any time to achieve more advanced functions.
The future life is like this. We just have to say something to the mobile phone at home. I am going to eat French snails at 12 noon today. The mobile phone will recommend it to the hotel, plan the route, arrange the time, all you have to do is go out of the house at the appointed time, board the car, and then get off the bus into the gate of the western restaurant. The car will automatically go to the gas station to refuel, return to the western restaurant door at the appointed time to pick you up.
This kind of life that looks like a sci-fi film may be realized in 10 years, and the future is very beautiful and at your fingertips.
The range of switch models comply with UL 61058-1 and CAN/CSA-22.2 No61058-1-05,for the United States and Canada.This series items are all high quality,and sale good.Many colors can be met.Especially,model KR2-11has high ampere,used wildly in the industrial area.
One of them,item KR2-11-201D is our hot sale item:
UL Approved Switch,New UL Approved Switch,UL Approved Micro Switch
Ningbo Kara Electronic Co.,Ltd. , https://www.kara-switch.com