More recently, innovation in the automotive industry has revolved around increasing engine power, then increasing efficiency, while at the same time improving aerodynamics, improving comfort levels and redesigning the appearance of vehicles. Now, hyperconnectivity and automation are driving the automotive industry into the future. Self-driving cars are the first thing that comes to mind when it comes to the car of the future, but the automotive industry tomorrow will be marked by far more than self-driving technology.
One of the main factors driving the transformation of cars is their connectivity, in other words, their connectivity, which paves the way for remote updates, predictive maintenance, improved driving safety and data protection from cyber threats. The cornerstone of connectivity, in turn, is the collection and storage of data.
Of course, the expansion of the networking capabilities of the car has made driving more enjoyable, but at the heart of this is the collection, processing and generation of a huge amount of data by the connected car. According to last year , within the next ten years, self-driving cars will learn to generate so much information that it will require more than 2 TB to store it, that is, much more space than it is now. And this is not the limit - as technology continues to develop, the figure will only grow. Based on this, equipment manufacturers must ask themselves how they can effectively respond to the needs of a significant increase in data in such an environment.
How will the architecture of unmanned vehicles develop?
Further improvement of functions such as data management generated by unmanned vehicles, object detection, map navigation and decision making depends largely on the success of machine learning and artificial intelligence models. The challenge for automakers is clear: the more advanced machine learning models become, the better the driving experience will be for users.
At the same time, the change in the architecture of unmanned vehicles takes place under the banner of optimization. Increasingly, manufacturers are opting for an extensive network of microcontrollers installed according to the needs of each specific application, preferring instead to install one large processor with serious computing power. It is this transition from multiple automotive microcontrollers (MCUs) to a single central MCU that is likely to be the most significant change in the architecture of cars of the future.
Transition of the data storage function from the car to the cloud
Data from unmanned vehicles can be stored both directly on board, if their prompt processing is necessary, and in the cloud, which is more suitable for in-depth analysis. The routing of data depends on its function: there is data that the driver needs immediately, for example, information from motion sensors or location data from a GPS system, in addition, based on them, the car manufacturer can draw important conclusions and, based on them, continue to work on improvement of the ADAS driver assistance system.
In a Wi-Fi coverage area, sending data to the cloud is economically feasible and technically simple, but if the car is in motion, the only option available may be a 4G (and eventually 5G) connection. And if the technical side of data transmission over a cellular network does not raise serious questions, its cost can be incredibly high. It is for this reason that many self-driving cars will have to be left for some time near the house or in some other place where they can be connected to Wi-Fi. This is a much cheaper option for uploading data to the cloud for further analysis and storage.
The role of 5G in the fate of connected cars
Existing 4G networks will continue to be the main communication channel for most applications, however, 5G technology can become a major catalyst for the further development of connected and autonomous cars, allowing them to almost instantly communicate with each other, with buildings and infrastructure (V2V, V2I, V2X ).
Autonomous cars cannot function without a network connection, and 5G is the key to fast connectivity and lower latency for the benefit of future drivers. Higher connection speeds will reduce the time it takes to receive the data collected by the vehicle, allowing the vehicle to respond almost immediately to sudden changes in traffic or weather conditions. The advent of 5G will also mark progress in the development of digital services for the driver and passengers, who will enjoy the ride even more and, accordingly, will increase the potential profit for these service providers.
Data Security: Who Holds the Keys?
It goes without saying that autonomous vehicles need to be protected by state-of-the-art cyber security. As stated in one , 84% of Automotive Engineers and IT Respondents expressed concern that automakers are failing to respond to ever-increasing cyber threats.
To ensure the integrity of the client and his personal data, all components of connected cars - from the hardware and software inside the car itself to the connection to the network and the cloud - must guarantee the highest level of security. The following are some of the measures that will help automakers ensure the security and integrity of the data used by self-driving cars.
- Cryptographic protection restricts access to encrypted data to a certain circle of people who know the valid "key".
- End-to-end security involves the implementation of a set of measures to detect a hacking attempt at every entry point into the data transmission line - from microsensors to 5G communication towers.
- The integrity of the collected data is an important factor and implies that the information received from the vehicles is kept intact until it is processed and converted into meaningful output. In case of damage to the converted data, this makes it possible to access the "raw" data and re-process them.
The Importance of Plan B
To perform all mission-critical tasks, the vehicle's central storage system must operate reliably. But how can automakers ensure that these goals are met if the system fails? One way to prevent accidents in the event of a failure of the main system would be to back up the data in a redundant data processing system, however, this option is incredibly expensive to implement.
Therefore, some engineers have gone the other way: they are working on creating backup systems for individual machine components involved in providing unmanned driving, in particular brakes, steering, sensors and computer chips. Thus, a second system appears in the car, which, without the obligatory backup of all data stored in the car, in the event of a critical equipment failure, can safely stop the car on the side of the road. Since not all functions are really vital (in an emergency, you can do without, for example, an air conditioner or radio), this approach, on the one hand, does not require the creation of a backup of non-critical data, which means a reduction in costs, and, on the other hand, all still provides insurance in case of system failure.
As the autonomous vehicle project develops, the entire evolution of transportation will be built around data. By adapting machine learning algorithms to process the vast amounts of data that autonomous vehicles depend on, and implementing robust and workable strategies to keep them safe and protected from external threats, manufacturers will at some point be able to design a car that is safe enough to drive on. digital roads of the future.
Source: habr.com