Communication is the new and necessary infrastructure to enable smart driving, said Qin Lihong, co-founder and president of NIO, adding that the move from 4G to 5G is a huge step forward and will certainly bring a revolutionary change in automated driving.
Qin Lihong explained his thoughts on 5G for smart driving in a forum presentation today, saying that as the car becomes smarter, it will go from being a complex machine with a mechanical architecture to being a mobile smart terminal.
The 4G network coverage will give the car more functions. For example, NIO is doing FOTA on a large scale right now to remotely push updates to iterate on the vehicle's underlying firmware and software performance systems, which can reduce the cycle time for a traditional car to solve a usage problem from months to seconds, he claims.
"Remote iteration (in automotive) is not a very common thing. We're used to using our iPhones and think remote upgrades are common. But in FOTA in the automotive sector, there are only two companies in the world that can do it, one is NIO and the other is Tesla."
Qin Lihong says this will even be a trend in the future because in 5 to 10 years from now, cars that don't support remote iteration won't be sold.
Qin Lihong believes that starting with remote iteration, the smart car industry will transition from auto-assisted driving to fully automated driving.
However, the vehicle's local computing is not yet realistic for the new energy vehicle industry, so it is more likely to be cloud computing.
This also determines whether it is automatically assisted driving or fully automated driving, are inseparable from four steps - perception, transmission, computational decision-making, and control. The transmission and computational decision process is closely related to communication.
And in the next stage of intelligent driving, communication is the new and necessary infrastructure to achieve intelligent driving.
"The move from 4G to 5G is definitely a huge one. But at a certain point, maybe 5G won't even be enough," Qin Lihong said.
And in the fully automated driving phase, Qin Lihong believes that roads as infrastructure also need to be intelligent, and smart cars become a key link in fully automated driving.
He used the example of a school of sardines in the sea, which consists of more than 200,000 sardines, but when they steer, they move in the same direction without colliding with each other, which is a bionic principle worth learning for future fully automated driving.
Qin Lihong said yes, the current network form is upgraded from telecommunications, there are base stations, computers, optical fiber, signal.
The next step is through neural networks, biomimicry, can we train and cultivate a new generation of networks that can learn and judge themselves.
This is an advanced network of vehicle-road collaboration, where when an accident or a situation occurs, the traffic sign and signal system of this road will signal 100 or 1000 cars driving on this road at the same time to do the same action of avoiding or stopping.
"(Automated driving) relies heavily on communications. Going from 4G to 5G will definitely bring a revolutionary change in automated driving. We can also expect 6G and 7G to bring faster, safer, and more secure communications until we push vehicles like cars to achieve a full degree of driverlessness," Qin Lihong said.
The following is the full text of the presentation:
I'm NIO co-founder and president Qin Lihong, who has been running the company since 2014 with founder and CEO William Li Bin.
Now I'm in charge of NIO's day-to-day general operations, spending more time on the branding and marketing services part of the business.
I'm very happy to be here today for such an interactive session.
I also hope that through today's limited time, combined with our smart grid development trends and NIO some rough practice, to tell you some of our thoughts on these areas.
As the car becomes smarter, it changes from a complex machine with a simple mechanical architecture to an intelligent terminal that can move.
At this point, things like 4G network coverage give the car more functionality.
I'll give you an example, like NIO is doing FOTA on a large scale, remotely to iterate on the underlying firmware and software performance systems of the vehicle.
This includes the ability for vehicle owners to talk directly to NOMI's robotic terminal through the vehicle's voice assistant and get your feedback on vehicle usage and quality straight to our quality engineer's back office in a matter of seconds.
In the traditional process, feedback on a quality or usage issue could be a months-long cycle.
In today's vehicles, we're able to respond in a matter of seconds. These are the advantages that 4G communication networks give us.
Let me go back to what I just said about FOTA.
There are two kinds of remote iterations, one called OTA and one called FOTA, with or without F. It matters a lot.
OTA is usually how we iterate on the central infotainment system of the vehicle, navigation, music, and interaction, and this FOTA usually represents some of the underlying components of the vehicle, including some gateway systems, operating systems, and so on.
For example, 2 months ago, 3 months ago, NIO remotely adjusted the hard and soft suspension of all the vehicle chassis through our version called NIO OS 2.6.1.
He adjusts the condition of the entire chassis, which is possible with FOTA.
NIO's FOTA has gone through 35 iterations since it went live in October 2018, bringing close to 100 new features to our users and fixing 200 to 300 issues.
And users' feedback on our vehicles simply needs to say "Hi NOMI, I want to report a problem" directly to NOMI, the intelligent robot in our in-car terminal and the various suggestions will reach our product and quality teams in seconds.
This user suggestion also forms part of the content of our next FOTA.
I think that through this kind of intelligent interoperability and very user-friendly interface, we can have a high-density and high-frequency interaction between the vehicle, the manufacturer, and the user, and work together to push the performance of the vehicle to a better level. This is a great scenario for an application.
I'd also like to give a preview of another major FOTA milestone that will soon be pushed out with version 2.7.0 of NIO's product.
This time we're doing it even more extreme, and we'll be recruiting a handful of users to be the first to experience it, looking for more scenarios that will hopefully be one step ahead of the remote iterations.
Many of you may think that remote iteration is not a very common thing, maybe we are used to using including Apple in smartphones and think that remote upgrades are common.
But what I want to tell you is that there are only two companies in the world that can do FOTA in the automotive industry, one is NIO and the other is Tesla.
Moreover, we have developed this system 100% by our own strength, which is a system that belongs to our own intellectual property rights and we own the whole system.
The OTA of many cars is not FOTA, it is done by a third party, and it can't touch the vehicle, and it can't reach the fundamental issues such as triboelectric management, power management, suspension management, and gateway management.
So this thing is a trend that has just emerged in the entire automotive industry, and I am very, very confident that in 5 to 10 years, but every car will be FOTA, and the cars that are not FOTA will not be sold, and this is called a trend.
When we talk about cars and car companies, we usually think of vehicles running on the road and factories on a grand scale.
In fact, today, there's an unseen front that I think we need to be aware of when we talk about the entire manufacturing and operations of the car business, and that's the cloud. Behind the entire physical vehicle and factory is a set of cloud-based systems based on IoT and mobile internet interactions.
To give you a simple example using NIO practices, for example, right now within our company, right in the user scenario, we have four network systems running in the background.
The first one we call it GRID, the grid, which is a path planning system.
When users drive an electric car, we are very concerned about the distance and complexity of the route, where can I charge and where can I battery swap.
In addition to the direct planning of the navigation path, we have to do a complementary energy planning, the complementary energy planning can not be easily completed is very critical.
We drive a fuel car rarely think about this problem because you go in any direction, there will always be a gas station in a short distance, this is the oil in the petrochemical industry for 70 years of work out of a network.
Electric vehicles as a new industry, whether it is the national power grid, like NIO such as host manufacturers, like emerging telecom third-party charging institutions, we do this thing is not a long time.
So if the abundance of resources is not enough, path planning becomes especially important.
If your friends are interested, you can look for a feature called the charging map in NIO's app, which is a full combination of third-party and NIO's own provision of charging battery swap stations, facilities, and navigation you want to go.
The computer will then tell you where you can stop to recharge.
At the charging site where you need to make a reservation, it will help you make a reservation in advance or remind you to make a reservation, it will help you calculate the most optimal route, help you estimate what time you might get tired, and then stop to recharge. People also take a break along the way, which is an optimal one.
This planning system is still a two-way system. When you use this map for planning, we also want users to write down their experiences and reviews, and more users' actual experiences will form a de facto review of the public charging stations.
This review density and reputation will also help the computer to determine for our users what is the most efficient, fastest, and lowest probability of failure for the public charging stations.
The charging map inside NIO's app is one such option. It is a public service product that helps NIO users as well as users of various brands of electric vehicles.
We also have a separate app called Jiadian (add power), which is what we used to call a gasoline vehicle, but now we call it an add power.
The more you use this add power app, the more intelligent this app system will be, and the GRID system behind it will help more users to plan more trips.
The second thing I want to introduce is that we have a system called OSS, which is an abbreviation of Operation Supporting System, which is such a system of operation assistance and operation support.
I remember at the end of 2017, our first NIO Day, William Li Bin, chairman of the board, said that every vehicle, every pile, every piece of equipment, every employee is highly networked.
When there's a local problem, it's the equivalent of you touching a network-wide response, and the entire company will mobilize the most optimal and nearest resources to help you.
This is a massive IoT system, which we call OSS, that connects all our service facilities, service vehicles, vehicle mechanics, service personnel, road boys, and our yard and back-office dispatch systems across the country.
Let me give you an example, if a user has a small accident somewhere in the East Fourth Ring Road and needs help, our whole network will calculate what level of a problem this is, which service station and technician is nearest to him, what method to use to rescue him, and how to notify the user.
At the same time, the rescue vehicle of this service station needs to be ready.
This is a case where an optimal service solution is accepted at the user end that feels flat and smooth.
But behind it is actually a result of the IoT system and algorithm support, which is not only connected to all of NIO's own resources but also further connected to the third-party resources we work with.
Next, we also want to initiate a practice of helping the masses. When a user needs help, most of the problems are probably nearest to another owner, who can provide help to him, and he will get there faster than our service personnel.
If we upgrade the system further, I think we will have a more agile and efficient OSS system where the owner helps the owner and the crowd helps the crowd.
The third system I'm going to introduce is called Athena, and we're naming it after the goddess of wisdom. Athena is our entire set of NIO energy, resource distribution, and planning back-office systems.
It serves two purposes: First, it can simulate, in a real-world environment, the changes in tenure, the changes in user trajectories, and the changes in geographic locations where users live.
How can we match our vehicles, stations, piles, people, and clouds with the same resource input?
In the second role, the Athena system can help our team to sandcastle the projection. Before we delivered our first NIO, we had already conducted hundreds of sandbox simulations in different cities and under different tenure conditions.
Let's say, we put in 10 battery swap stations in Beijing in the first year, how these 10 battery swap stations should be arranged optimally. If the number of users and the distribution of A, B, C, D, E are different, how will it be distributed to reach the optimal solution?
After we look for this intersection, we get to the business work on the ground. Because the optimal place in your mind you may not be able to get, you may not be able to negotiate down, so there are optimal solutions, suboptimal solutions, and so on, and when one point falls to a suboptimal answer, the other points have to change accordingly to get the whole grid layout to the optimal state.
This Athena system is all computed dynamically, and it's constantly guiding us to the best plan, which is very fun.
It's very similar to a game of getting resources, build a base, you have an infinite number of options, and in the end, you decide on a more realistic one and put it to work.
Finally, I'd like to introduce you to our Battery Shield, which is a system for electromagnetic safety management and remote monitoring.
As we all know, electric vehicles are new things that represent the direction of technology, and inevitably, short Translated with www.