In February 2019, Musk blew a big bubble to the media. He said that Tesla will soon launch a fully autonomous driving function by the end of the year, and if given another year (that is, 2020), Tesla will dare to let the driver sleep in the driver's seat.
Musk also specifically described a scenario in which Tesla cars in 2020 will "can find you in the parking lot, pick you up, and take you to your destination."
Everyone knows what happened next. Musk repeats this "joke" almost every year, but fully autonomous driving still seems to be far away from us. Even when many people hear this word, they can’t help but label it a “liar”.
However, although fully autonomous driving is still far away, the most difficult urban ADAS is infinitely closer to us. Especially at the upcoming Chengdu Auto Show, this technology is about to shine.
Why do you say that? Let Kung Fu Motors take everyone to see it together.
(1) Encountering a bottleneck, autonomous driving is blocked?
Whether it is Tesla, Google, Baidu, Huawei, or Xpeng, they have all encountered bottlenecks on the road to autonomous driving in recent years.
The first thing to be defeated should be the "Internet of Vehicles", which realizes the interconnection between cars and roads, cars and cars, cars and people, and cars and sensors by connecting everything to the Internet.
In theory, this should be regarded as the ultimate solution for autonomous driving. The car only needs to be connected to the network like a mobile phone. It is clear at a glance where each car is going and which road it wants to take. It is hoped that the speed and direction of the vehicle can be determined, and the commuting efficiency will become infinitely higher.
But the ideal is full, but the reality is extremely skinny. Let’s not talk about how expensive it is to build roads with Internet of Vehicles functions. Even if all new cars are connected to the Internet, what will happen to those old cars that have already been on the market? The number of these cars can be counted in hundreds of millions, and they are outside the network. Not to mention how much computing power is needed to implement a large network that can coordinate all transportation needs.
The second one to be defeated should be the so-called high-precision map route, because this technical route requires collecting data at any time, traveling all over the country to ensure the map, and at the same time, it must be updated in real time to ensure safety during driving. As a result, the cost of high-precision maps is too high, and the cost will not be reduced due to heavy investment in the early stage, so naturally there is no promotion value.
The third path that all major car companies are insisting on is the strong perception path. It senses things around it through cameras and radar (Tesla has recently added radar), and then controls it through the AI model to complete driving. In theory, as long as AI thinking is strong enough, with the excellent "vision" of the cameras, it can drive like a human and be more precise than human control. But this road is also full of difficulties. The biggest difficulty is that no one knows how much computing power an AI that thinks like a human will need.
Why are major car companies rushing to claim that they want to achieve autonomous driving around 2020, but now that the technology has iterated a lot, they have remained silent? In fact, the more you know, the more difficult it is.
(2) The commuting mode emerges and becomes a new hot spot?
Since autonomous driving will most likely not be realized in a short time, does ADAS technology have no value? In fact, it is not the case. In fact, high-end intelligent driving technology is gradually changing our lives.
The most typical one is parking. In the early days, most people relied on "looking in the rearview mirror" and looking for a few key "points" in each direction when reversing to complete reversing. With the popularity of 360, reversing has become very easy for most people. The recent emergence of various automatic parking vehicles, which can park at various positions and at various weird angles, is even more exciting for everyone. If those born in the 1990s are the first drivers who can’t look at the rearview mirror to reverse, those born in the 2000s may not even be able to see the reversing image.
The same is true for high-speed commuting. The earliest cruise control was very difficult to use, and it could only be cruised for a short period of time unless the road conditions were excellent. After the emergence of ACC, the vehicle speed can be adjusted by itself, and the practicality will increase a lot. In particular, some car companies have combined lane keeping technology and basically have piloting functions. Nowadays, major car companies have done a very good job in high-speed NOA. After setting the path, you can cruise to the end, which is really enjoyable.
The "commute mode" is another innovation. Users do not need to wait for the so-called high-precision map to be opened, nor do they need to wait for the NOA push of a certain city. They can directly start the system to learn on their own based on their daily commuting routes.
In this way, high-end intelligent driving directly changes from "futures" to "spot". You can use it as soon as you get it, and it can also be "tailor-made". After all, most of us actually only have a few routes for daily commuting. This is equivalent to meeting more than 50% of autonomous driving needs at a very small cost. The person who came up with this technical solution is indeed a genius.
Of course, we can also look forward to it. If everyone's training route is opened, wouldn't there be countless collectors of high-precision maps, and they will be updated in real time. Is autonomous driving really far away from us?
(3) Commuting mode, the biggest highlight of ADAS?
Therefore, at the next Chengdu Auto Show, the "commute mode" solutions launched by major car companies will be the biggest attraction in the field of intelligent driving. Let's put it this way, if a car company says that it has achieved autonomous driving, it will most likely have to cut leeks. If a certain car company says that its commuting mode has been opened up, then it is most likely worth taking a look.
Compared with high-precision maps that limit ODD areas to fixed directions, the commuting mode is obviously more suitable for daily traffic in the city. In fact, the essence of the two is the same, they both limit the geographical scope of the function. But the difference is that we cannot control the range of high-precision map activation, and it is currently concentrated in some fixed areas in first- and second-tier cities. The commuting mode changes the function activation range to the range of our daily use of the car. Is there anything more considerate than this?
Moreover, the essence of autonomous driving is actually two directions, driving and parking. Nowadays, parking has become more and more urban, and driving on highways and elevated roads has gradually matured. The most difficult part is commuting in the city. The road conditions we face are too complicated. Nowadays, commuting mode alleviates this very well.
More importantly, the collection of high-precision maps is often one-time (no one has the resources to collect it repeatedly). That means that certain emergencies that occur are often captured, such as a bird that suddenly flies by, or a leaf that floats by, which will be remembered by the map. The commuting mode does not have this problem. After all, its calculation model comes from daily driving and can be continuously improved, iterated, and upgraded to become more and more useful.
(4) Kung Fu shooting
The term "commute mode" was first proposed by Xiaopeng. However, Ideal, Weilai, Wenjie, and SAIC have all followed up almost immediately. At present, almost all domestic smart driving car companies have invested in this track.
But the most exaggerated one is the Baojun Yunduo. As a 100,000-class model, it is also equipped with three major functions: memory parking, tracking reversing, and commuting mode.
This also shows that the commuting mode has the foundation for rapid civilianization. It is easy to use and not expensive, which means that its promotion will be very, very fast. What will happen at the Chengdu Auto Show?
Previous article:Another global leader enters the lithium iron phosphate battery market!
Next article:Solving the problem of energy replenishment: How far is wireless car charging from us?
- Popular Resources
- Popular amplifiers
- Car key in the left hand, liveness detection radar in the right hand, UWB is imperative for cars!
- After a decade of rapid development, domestic CIS has entered the market
- Aegis Dagger Battery + Thor EM-i Super Hybrid, Geely New Energy has thrown out two "king bombs"
- A brief discussion on functional safety - fault, error, and failure
- In the smart car 2.0 cycle, these core industry chains are facing major opportunities!
- The United States and Japan are developing new batteries. CATL faces challenges? How should China's new energy battery industry respond?
- Murata launches high-precision 6-axis inertial sensor for automobiles
- Ford patents pre-charge alarm to help save costs and respond to emergencies
- New real-time microcontroller system from Texas Instruments enables smarter processing in automotive and industrial applications
- Innolux's intelligent steer-by-wire solution makes cars smarter and safer
- 8051 MCU - Parity Check
- How to efficiently balance the sensitivity of tactile sensing interfaces
- What should I do if the servo motor shakes? What causes the servo motor to shake quickly?
- 【Brushless Motor】Analysis of three-phase BLDC motor and sharing of two popular development boards
- Midea Industrial Technology's subsidiaries Clou Electronics and Hekang New Energy jointly appeared at the Munich Battery Energy Storage Exhibition and Solar Energy Exhibition
- Guoxin Sichen | Application of ferroelectric memory PB85RS2MC in power battery management, with a capacity of 2M
- Analysis of common faults of frequency converter
- In a head-on competition with Qualcomm, what kind of cockpit products has Intel come up with?
- Dalian Rongke's all-vanadium liquid flow battery energy storage equipment industrialization project has entered the sprint stage before production
- Allegro MicroSystems Introduces Advanced Magnetic and Inductive Position Sensing Solutions at Electronica 2024
- Car key in the left hand, liveness detection radar in the right hand, UWB is imperative for cars!
- After a decade of rapid development, domestic CIS has entered the market
- Aegis Dagger Battery + Thor EM-i Super Hybrid, Geely New Energy has thrown out two "king bombs"
- A brief discussion on functional safety - fault, error, and failure
- In the smart car 2.0 cycle, these core industry chains are facing major opportunities!
- The United States and Japan are developing new batteries. CATL faces challenges? How should China's new energy battery industry respond?
- Murata launches high-precision 6-axis inertial sensor for automobiles
- Ford patents pre-charge alarm to help save costs and respond to emergencies
- New real-time microcontroller system from Texas Instruments enables smarter processing in automotive and industrial applications
- Linux system boot card production and system hardening
- RH850 development method advice
- [Repost] An article summarizes eight IO port modes of STM32 microcontrollers
- [Fudan Micro FM33LG0 Series Development Board Review] Connecting the TJA1050 CAN Communication Module
- EEWORLD University Hall----Live Replay: STMicroelectronics Home Appliance Three-Phase Motor Control Solution
- Date in spring + buy flowers to decorate the room
- [Raspberry Pi 3B+ Review] Install libjpeg library & drive USB camera
- C language BCD code to decimal (hexadecimal HEX) program
- From ERM to X-axis LRA, TI provides you with a comprehensive Haptics Solution
- Single-cell Li-ion battery fuel gauge and low cutoff current charging IC