The future of autonomous driving connectivity – Quantum entanglement or 6G?

 The development of connected car technology accelerated digital disruption in the automotive industry. Verified Market Research valued the connected car market at USD 72.68 billion in 2019 and projected its value to reach USD 215.23 billion by 2027. Along with the rapid growth of this market’s worth, we observe the constant development of new customer-centric services that goes far beyond driving experience.

While the development of connected car technology created a demand for connectivity solutions and drive-assistance systems, companies willing to build their position in this market have to face some significant challenges. This article is the first one of the mini-series that guides you through the main obstacles with building software for connected cars. We start with the basics of a connected vehicle, then dive into the details of prototyping and providing production-ready solutions. Finally, we analyze and predict the future of verticals associated with automotive-rental car enterprises, insurers, and mobility providers.

This series provides you with hands-on knowledge based on our experience in developing production-grade and cutting-edge software for the leading automotive and car rental enterprises. We share our insights and pointers to overcome recurring issues that happen to every software development team working with these technologies.

What is a Connected Car?

A Connected Car is  a vehicle that can communicate bidirectionally with other systems outside the car , such as infrastructure, other vehicles, or home/office. Connected cars belong to the expanding environment of devices that comprise the Internet of Things landscape. As well as all devices that are connected to the internet, some functions of a vehicle can be managed remotely.

Along with that, IoT devices are valuable resources of data and information that enable further development of associated services. And while most car owners would describe it as the mobile application paired with a car that allows users to check the fuel level, open/close doors, control air conditioning, and, in some cases, start the ignition, this technology goes much further.

V2C - Vehicle to Cloud

Let’s focus on some real-case scenarios to showcase the capabilities of connected car technology. If a car is connected, it may also have a sat-nav system with a traffic monitoring feature that can alert a driver if there is a traffic jam in front of them and suggest an alternative route. Or maybe there is a storm at the upcoming route and navigation can warn the driver. How does it work?

That is mostly possible thanks to what we call V2C - Vehicle to Cloud communication. Utilizing the fact that a car is connected, and it is sending and gathering data, a driver may also try to find it, in case it was stolen. Telematics data is also helpful to understand the reasons behind an accident on the road - we can analyze what happened before the accident and what may have led to the event. The data can be also used for predictive maintenance, even if the rules managing the dates are changing dynamically.

While this seems just like a nice-to-have feature for the drivers, it allows car manufacturers to provide an extensive set of subscription-based features and functionalities for the end-users. The availability of services may depend on the current car state - location, temperature, and technical availability. As an example: during the winter, if the car is equipped with heated seats and the temperature drops under 0 Celsius, but the subscription for this feature expires, the infotainment can propose to buy the new one - which is more tempting when the user is at this time cold.

V2I - Vehicle to Infrastructure

A vehicle equipped with connected car technology is not limited to communicating only with the cloud. Such a car is capable of exchanging data and information with road infrastructure, and this functionality is called V2I - Vehicle to Infrastructure communication. A car processes information from infrastructure components - road signs, lane markings, traffic lights to support the driving experience by suggesting decision makings. In the next steps, V2I can provide drivers with information about traffic jams and free parking spots.

Currently, in Stuttgart, Germany, the city’s infrastructure provides the data live traffic lights data for vehicle manufacturers, so drivers can see not just what light is on, but how long they have to wait for the red light to switch to green again. This part of connected car technology can rapidly develop with the utilization of wireless communication and the digitalization of road infrastructure.

V2V - Vehicle to Vehicle

Another highly valuable type of communication provided by connected car technology is V2V - Vehicle to Vehicle. By developing an environment in which numerous cars are able to wirelessly exchange data, the automotive industry offers a new experience - every vehicle can use the information provided by a car belonging to the network, which leads to more effective communication covering traffic, car parking, alternative routes, issues on the road, or even some worth-seeing spots.

It may also significantly increase safety on the road, when one car notifies another that drives a few hundred meters behind him that it just had a hard breaking or that the road surface is slippery, using the information from ABS, ESP, or TC systems. That has not just an informational value but is also used for Adaptive Cruise Control or Travel Assist systems and reduces the speed of vehicles automatically increasing the safety of the travelers. V2V communication makes use of network and scale effects - the more users have connected to the network, the more helpful and complete information the network provides.

The list of use cases for connected car technology is only limited by our imagination but is excelling rapidly as many teams are joining the movement aiming to transform the way we travel and communicate. The Connected Car revolution leads to many changes and impacts both user experience and business models of the associated industries.

How connected car technology impacts business models of the automotive industry

Connected cars bring innovative solutions to the whole environment comprising the automotive landscape. Original Equipment Manufacturers (OEMs) have gained new revenue streams. Now vehicles allow their users to access stores and purchase numerous features and associated services that enhance customer experience, such as infotainment systems. By delivering aftermarket services directly to a car, the automotive industry monetizes new channels. Furthermore, these systems enable automakers to deliver advertisements, which become an increasing source of revenue.

 The development of new technology in automotive creates a similar change as we observed in the mobile phone market. When smartphones equipped with operating systems had become a new normal, significantly increased the number of new apps that now allow their users to manage numerous services and tasks using the device.

But it is just an introduction to numerous business opportunities provided by connected cars. Since data has become a new competitive advantage that fuels the digital economy, collecting and distributing data about user behavior and vehicle performance is seen as highly profitable, especially when taking into account the potential interest of insurers providers.

Assembled data while used properly gives OEMs powerful insights into customer behavior that should lead to the rapid growth of new technologies and products improving customer experiences, such as predictive maintenance or fleet management.

The architecture behind connected car technology

Automotive companies utilize data from vehicle sensors and allow 3rd party providers to access their systems through dedicated API layers. Let’s dive into such architecture.

 High-Level Architecture

System components

Digital Twin in automotive

A digital twin is a virtual replica and software representation of a product, system, or process. This concept is being adopted and developed in the automotive industry, as carmakers utilize its powerful capabilities to increase customer satisfaction, improve the way they develop vehicles and their systems, and innovate. A digital twin empowers automotive companies to collect various information from numerous sensors, as this tool allows to capture operational and behavioral data generated by a vehicle. Equipped with these insights, the leading automotive enterprises work on enhancing performance and customize user experience, but meanwhile, they have to tackle significant challenges.

First of all, getting data from vehicles is problematic. Hardware built-in vehicles have particular limits, which leads to reduced capabilities in providing software. Unlike software, once shipped hardware cannot be easily adjusted to the changing conditions and works for several years at least. Furthermore, while willing to deliver a customer-centric experience, automakers still have to protect their users from numerous threats. To protect vehicles from denial of service attacks, vehicles can throttle the number of requests. Overall, it’s a good idea but can have a terrible impact when multiple applications are trying to get data from vehicles, e.g., in the rental domain. This complex problem can be simply solved by Digital Twin. It can expose data to all applications without them needing to connect to the vehicle by simply gathering all real-time vehicle data in the cloud.

Implementation of this pattern is possible by using NoSQL databases like MongoDB or Cassandra and reliable communication layers, examples are described below. Digital Twin may be implemented in two possible ways, uni- or bidirectional.

Unidirectional Digital Twin

Unidirectional Digital Twin is saving only values received from the vehicle, in case of conflict it resolves the situation based on event timestamp. However, it doesn’t mean that the event causing the conflict is discarded and lost, usually every event is sent to the data platform. The data platform is a useful concept for data analysis and became handy when implementing complex use cases like analyzing driver habits.

Bidirectional Digital Twin

The Bidirectional Digital Twin design is based on the concept of the current and desired state. The vehicle is reporting the current state to the platform, and on the other hand, the platform is trying to change the state in the vehicle to the desired value. In this situation, in case of conflict, not only the timestamp matters as some operations from the cloud may not be applied to the vehicle in every state, eg., the engine can’t be disabled when the vehicle is moving.

However, meeting the goal of developing a Digital Twin may be tricky though as it all depends on the OEM and provided API. Sometimes it doesn’t expose enough properties or doesn’t provide real-time updates. In such cases, it may be even impossible to implement this pattern.

API

At first, designing a Connected Car API isn’t different from designing an API for any other backend system. It should start with an in-depth analysis of a domain, in this case, automotive. Then user stories should be written down, and with that, the development team should be able to find common parts and requirements to be able to determine the most suitable communication protocol. There are a lot of possible solutions to choose from. There are several reliable and high-traffic oriented message brokers like Kafka or hosted solutions AWS Kinesis. However, the simplest solution based on HTTP can also handle the most complex cases when used with Server-Sent Events or WebSockets. When designing API for mobile applications, we should also consider implementing push notifications for a better user experience.

When designing API in the IoT ecosystem, you can’t rely too much on your connection with edge devices. There are a lot of connectivity challenges, for example, a weak cellular range. You can’t guarantee when your command to a car will be delivered, and if a car will respond in milliseconds or even at all. One of the best patterns here is to provide the asynchronous API. It doesn’t matter on which layer you’re building your software if it’s a connector between vehicle and cloud or a system communicating with the vehicle’s API provider. Asynchronous API allows you to limit your resource consumption and avoid timeouts that leave systems in an unknown state. It’s a good practice to include a connector, the logic which handles all connection flaws. Well designed and developed connectors should be responsible for retries, throttling, batching, and caching of request and response.

OEM’s are now implementing a unified API concept that enables its customers to communicate with their cars through the cloud at the same quality level as when they use direct connections (for example using Wi-Fi). This means that the developer sees no difference in communicating with the car directly or using the cloud. What‘s also worth noting: the unified API works well with the Digital Twin concept, which leads to cuts in communication with the vehicle as third-party apps are able to connect with the services in the cloud instead of communicating directly with an in-car software component.

What’s next for connected car technology

Once the challenges become tackled, connected vehicles provide automakers and adjacent industries with a chance to establish beneficial co-operations, build new revenue streams, or even create completely new business models. The possibilities delivered thanks to over-the-air communication (OTA) allowing to send fixes, updates, and upgrades to already sold cars, provide new monetization channels, and sustain customer relationships.

As previously mentioned, the global connected car market is projected to reach USD 215.23 billion by 2027. To acquire shares in this market, automotive companies are determined to adjust their processes and operations. Among key factors that impact the development of connected car technology, we can point out a few crucial. The average lifecycle of a car is about 10 years. Today, automakers make decisions regarding connected cars that will go into production two to four years from now. For the cellular connectivity strategy to remain relevant over 12 to 15 years, significant challenges and assumptions need to be collaboratively addressed by OEMs, telematics control unit suppliers, and service providers.

 Automakers must manage software in the field reliably, cost-efficiently, and, most importantly, securely – not just patch fixes, but also continually upgrade and enhance the functionality. The availability of OTA updates reduces the burden on dealerships and certified repair centers but requires better and more extensive testing, as the breakage of critical features is not an option.

 Cellular solutions need to be agile to be compatible with emerging network technologies over the vehicle lifetime, e.g., 5G to be the industry standard in the next few years. The chosen solution must deliver reliable, seamless, uninterrupted coverage in all countries and markets where the vehicles are sold and driven.

 Solution developers must offer scalable, cost-effective ways to develop upgradeable software that can be universally deployed across technologies, hardware, and chipsets. A huge focus must be put on testing the changes automatically on both the cloud platform side and the vehicle side.

As Connected Vehicles proliferate, the auto industry will need to adapt and transform itself into the growing technological dependency. OEMs and Tier-1 manufacturers must partner with technology specialists to thrive in an era of software-defined vehicles. As connectivity requires skills and capabilities outside of the OEMs’ domain, automakers will necessarily have to be software developers. An open platform environment will go a long way to encourage external developers to design apps for vehicle connectivity platforms.

Generative AI is becoming a major player in automotive innovation. The market is already valued at  USD 480.22 million in 2024 , and it’s expected to grow to USD 3,900.03 million by 2034, with a steady annual growth rate of 23.3%. Moreover, by 2025, the global automobile sector will invest $11.1 billion in cognitive and AI technologies. These numbers show how quickly the industry is picking up on this technology’s potential.

GenAI is making its mark across various areas. From manufacturing optimization to autonomous driving, its impact is undeniable. Predictive maintenance systems identify issues early, AI-powered tools optimize vehicle development, and talking to  in-car assistants is starting to feel like a scene out of a sci-fi movie.

Speaking of sci-fi, pop culture has always loved the idea of talking cars. There is K.I.T.T. (Knight Industries Two Thousand), of course, but also all Transformers and tons of cartoons, starting with Lightning McQueen. Is it just pure fiction? Not at all (except McQueen, for many reasons 😊)! Early attempts at smarter cars started with examples like a 2004 Honda offering voice-controlled navigation and Ford’s 2007 infotainment system. Fast forward to now, and we have a VW Golf with a GPT-based assistant that’s more conversational than ever.

But honestly, the most resourceful one is K.I.T.T. – it activates all onboard systems, diagnoses itself, and uses company resources (there is an episode when K.I.T.T. withdraws money from the company bank account using an ATM). In 1982, when the show first aired, it was just pure science fiction. But what about now? Is it more science or fiction? With  Generative AI growing rapidly in automotive, we have to revisit that question.

Let’s break it down!

Prerequisites

Let’s assume we would like to create  a solution-oriented chatbot connected with a car. By “solution-oriented,” I mean one that is really useful, able not only to change the attractive interior lighting but also to truly solve owners’ issues.

The idea is to use Generative AI, a large language model with its abilities in reasoning, problem-solving, and language processing.

 Therefore, the first question is – where should the model be planted – in the cloud or a car?

For the first option, you need a constant Internet connection (which is usually not guaranteed in cars). In contrast, the second option typically involves a smaller and less versatile model, and you still need a lot of resources (hardware, power) to run it. The truth lies, as usual, in between (cloud model if available, local one otherwise), but today we’ll focus on the cloud model only.

 The next step is to consider the user-facing layer. The perfect one is integrated into the car, isn’t it? Well, in most cases, yes, but there are some drawbacks.

The first issue is user-oriented – if you want to interact with your car when being outside of it, your mobile phone is probably the most convenient option (or a smartwatch, like Michael from Knight Rider). Also, infotainment systems are comprehensively tested and usually heavily sealed into cars, so introducing such a bot is very time-consuming. Therefore, the mobile phone is our choice.

We don’t want to focus on this application today, however. Depending on the target operating system, it probably should use speech-to-text recognition and text-to-speech generation and stream data both ways for a better user experience.

The core part is the chatbot backend – a regular application connecting the frontend and the LLM. It should be able to call external APIs and use two sources of knowledge – live car data and company-owned data sources.

Basics

Let’s gather the components. There is a customer-facing layer – the mobile application; then there is our main backend application, the LLM, of course, and some services to provide data and functionalities.

The diagram above is conceptual, of course. The backend is probably cloud-hosted, too, and cloud services linked to car services form the essence of the “connected cars” pattern.

 The main concept for the application is “tool calling” – the LLM ability to call predefined functions with structuralized arguments. That’s why the backend is surrounded by different services. In a perfect world, those should be separated microservices designed for different use cases. However, this architecture is not scenario-based. There is no “if-else-if” ladder or so. The LLM determines how to utilize the tools based on its own decision-making process.

The sample conversation schema might look like the one presented below.

As you can see, the chatbot service calls the LLM, and the LLM returns command “call function A.” Then, the service calls the function and returns the response to the LLM (not the user!).

This approach is very flexible as functions (a.k.a. tools) might execute actions and return useful data. Also, the LLM may decide to use a function based on another function result. In the case above, it can, for example, use one function to check the climate control system status and discover that it’s running in the “eco mode”. Then, it might decide to call the “set mode” function with the argument “max AC” to change the mode. After that, the LLM can return an answer to the user with a message like “It should be fixed now”.

To build such an application, all you need to call the LLM like that (OpenAI GPT-4o example):

{
 "model": "gpt-4o",
 "messages": [
   {
     "role": "user",
     "content": "My AC is ineffective! Fix it!"
   }
 ],
 "tools": [
   {
     "type": "function",
     "function": {
       "name": "get AC status",
       "description": "Return current status of the climate control system"
     }
   },
   {
     "type": "function",
     "function": {
       "name": "set AC mode",
       "description": "Sets up the specified mode for the climate control system",
       "parameters": {
         "type": "object",
         "properties": {
           "mode": {
             "type": "string",
             "description": "Desired mode",
             "enum": ["ECO", “NORMAL”, "MAX AC"]
           }
         },
         "required": ["mode"]
       }
     }
   }
 ],
 "tool_choice": "auto"
}

As you can see, the response schema does not bother us here – the assumption is that the LLM is able to understand any reasonable response.

Dive

The subtitle should be a “deep dive”, but honestly, we’re just scratching the surface today. Nevertheless, let’s focus a little bit more.

So far, we have the user-facing application and the backend service. Now, let’s make it useful.

The AC example mentioned above is perfectly valid, but how can it be achieved? Let’s say there is an API for interaction with the AC in the car. It’s typical for all PHEVs and EVs and available for some HEVs, too, when you can turn on your AC remotely via the mobile app. However, the real value lies in the connected car

There is no IP address of the car hardcoded in the application. Usually, there is a digital twin in the cloud (a cloud service that represents the car). The application calls the twin, and the twin notifies the vehicle. There should also be some pub/sub queue in between to handle connectivity tier disruptions. Also, the security layer is extremely important. We don’t want anybody even to play the radio at max volume during a quiet night ride, not to mention turning off the lights or engaging breaks.

 Which brings us to the list of possible actions.

Let’s assume all systems in the car are somehow connected, maybe using a common bus or a more modern ethernet-like network. Still, some executors, such as brakes, should be isolated from the system.

So, there is no “brake API” to stop a car. However, it may be beneficial for mechanics to execute some "dangerous" actions programmatically, e.g., to increase the pressure in the braking system without actually pressing the pedal. If this is the case, such functionalities should be accessible exclusively through a local connection without the need for digital twin integration. Therefore, we can assume there are two systems in the car – local and cloud-integrated, no matter the nature of the isolation (physical, network, or software). Let’s focus on the  connected car aspect.

I believe the system should be able to change the vehicle settings, even if there is a risk that the driver could be surprised by an unauthorized change in the steering feel while taking a turn. This way, the chatbot might be useful and reduce support load by adjusting car settings based on the user's preferences. To avoid misusage, we can instruct the chatbot by prompt engineering to confirm each change with the user before execution and, of course, implement best-in-class security for all components. We can also allow certain operations only if the car is parked.

 Which brings us back to the list of possible actions.

For the sake of this article, let’s assume the chatbot can change various car settings. Examples include:

•  Climate control settings
•  Driver assistant sensitivity and specific functions toggles
•  Navigation System settings, like route type or other functions toggles
•  360 camera system settings, like brightness adjustment
•  Sound system settings like equalizer
•  Wiper settings
•  Notifications settings
•  Active steering system settings

This list is not complete, and the best thing is – it doesn’t need to be, as adding new functions (tool definition + API availability) might be a part of the future system OVA update.

What about reading real-time data? Should we connect to the car directly and read the status? Let’s leave this option for another article 😉 and focus on communication via the cloud.

 There are two possibilities.

We can provide more tools to get data per source/component (a reminder – LLM decides to call for data, which then triggers an API call, and the LLM processes the received response). Alternatively, we could implement a single tool, “get vehicle data,” that collects and merges all data available from all data sources.

For the latter approach, two ways are available – do we really need a tool? Maybe we should inject the current state into each conversation, as it’s probably beneficial to have the current state anyway to solve all cases?

Let me give the standard consultant reply to those questions.

 It depends.

More data in the context means extended response time and a higher bill. Also, some cases don’t need the entire context or don’t need it at all. On the other hand, if you let the LLM decide which part of the context is necessary (which function to call), it will also affect the time and cost.

The next part is the “cost” of collecting data. Some data sources might be slow or consume valuable in-car resources to calculate (maybe not appropriate) data.

My advice would be to start collecting all data in the background during session creation and attach it to the session progressively when ready. Additionally, provide the LLM with the capability to wait for data if it has not yet been supplied. This should be implemented as a function to handle instances where the user requests data-related action, but the necessary data is not yet available

The second part is splitting the data. You can use a much smaller LLM to decide which context data is necessary to answer the current question and remove unnecessary information to keep it concise.

The last part is company data. All LLM models are trained based on thousands of manuals available over the Internet, but this is usually not sufficient. Your chatbot should have the car manual at its fingertips, so the general knowledge is not enough. Therefore, you need to provide it. You can execute the  RAG procedure and add some content to the context for each prompt, but my advice would be to hide it behind a function, too. Then, the LLM may decide when to call it to get some extra information if needed.

Use cases of Generative AI for connected cars

OK, we have all the components in place – there is the frontend, the backend, the model, and the toolbox. Can it be useful? As you would expect!

In the examples below, I will skip the part where the LLM commands the backend to perform an action, and the backend calls an API. I will also omit the process of the API response traveling back through the backend to the LLM. Additionally, I will assume that no new data is added to the context, so the LLM must decide which data it needs. Besides that, make sure that your context contains some system prompts to better explain all possibilities to the assistant.

Here are some examples.

                   User:         I think my Lane Keeping Assist activates too frequently. Can you check it?    
                Assistant:               [get driver assistance settings]    
                Assistant:               [diagnose sensors]    
                Assistant:               Everything looks good, but I can see that the Lane Keeping Assist mode is set to “fast”. Do you want me to lower the setting to “medium”?    
         User:         Yes, please.    
                Assistant:               [change LKA settings]    
                Assistant:               It should be better now.          

In the example above, the user has a problem with one of the driver assistance components. The system proposes and, after confirmation, executes settings changes, so the driver doesn’t need to find a proper setting in the on-board computer.

Check another case.

                   User:         Whenever I’m parking the car in my garage, I can barely see anything in my rear camera. There is not much light there, but maybe you help me in some other way?    
                Assistant:               [get camera settings]    
                Assistant:               I can increase the brightness of the camera, but then it might be too bright for other places. What do you think?    
         User:         Yes, please do that.    
                Assistant:               [change settings]    
                Assistant:               Here you are; let me know if you want me to roll back the change.          

Usually, there is no brightness setting available for the user. To adjust it, the user would need to schedule a service department meeting, spend some time, and pay for the facility. The assistant may do it in a few seconds for free.

                   User:         There are so many notifications in the car, so I can’t hear the important ones!    
                Assistant:               [check last notifications]    
                Assistant:               If you find some of my notifications unnecessary, I can mute some of them. What do you think?    
         User:         Yes, please turn off the speeding avoidance warning.    
                Assistant:               I can’t turn it off due to legal requirements, but I can replace the sound with vibration. feedback on the acceleration pedal. Do you want me to do that?    
         User:         Sounds great!    
                Assistant:               [adjust settings]    
                Assistant:               It’s done. I hope you’ll be less distracted.          

In this case, the assistant solves the issue of annoying notifications. Note the alternative solution suggestion and knowledge about the legal aspect. We can achieve it with proper function descriptions.

                   User:         It seems my car can’t connect to the Internet.    
                Assistant:               [check the connection] -> no connection to the car from the cloud service    
                Assistant:               It seems your car is disconnected. Can you please check the cellular network signal icon on the infotainment main screen?    
         User:         I can see the signal is good.    
                Assistant:               [check last logs]    
                Assistant:               I can see there were some issues with your connection before. Please try to open the settings in the car infotainment and restart the communication module. I’ll guide you through the process.    
         User:         OK, what should I do?    
                Assistant:               [check user manual]    
                Assistant:               Go to “Settings”, then “Connectivity”, then “Cellular” and then click “Restart module”. Confirm when asked.    
         User:         OK    
                Assistant:               [check the connection]    
                Assistant:               I can see the connection is restored. Can you confirm that? If not, I can schedule a service meeting for you.    
         User:         No need; it works now!          

In the last example, the Assistant guides the user using logs, the manual, and after checking real-time conditions. Without the LLM, the user would need to visit the dealer or at least use online support assistance.

Final words

With modern technology, we can bring some science fiction concepts to real life. Maybe WARP speed is not possible yet, but at least a car itself can give us solutions to problems recently solved only in much more expensive ways.

The beauty of it is – it’s really simple. Of course, there is a user-facing application to be made, the security to be ensured, and a bunch of functions to be implemented, but with modern connected cars, those elements are in place anyway. The tricky, novel part is a very simple application that uses the LLM as the brain power of the system.

As you can see the “  Attention is all you need ” paper that started the LLM revolution has allowed humanity to bring to life concepts present in our culture for decades. On the other hand, would this article have been ever written if its authors hadn’t watched the K.I.T.T. in their childhood? We will never know.

 Automotive is transforming into a hyper-connected, software-driven industry that goes far beyond the driving experience. How to build applications in such an innovative environment? What are the main challenges of providing software for connected cars and how to deal with them? Let’s dive into the process of utilizing the capabilities of the cloud to move automotive forward.

People have always aimed for the clouds. From Icarus in Greek mythology, first airplanes and spaceships to dreams about flying cars – our culture and history of technology development express a strong desire to go beyond our limits. Although the vision from Back to the Future and other Sci-Fi movies didn’t come true and our cars cannot be used as flying vehicles, our cars actually are in the cloud.

Meanwhile, the idea of the Internet of Things came true; our  devices are connected to the Internet . We have smartphones, smartwatches, smart homes and, as it turns out, smart cars. We are able to communicate with them to gather data or even remotely control them. The possibilities are only limited by hardware, but even it is constantly improving to follow the pace of rapid changes triggered by software development.

Offerings on the automotive market are developing rapidly with numerous features and promised experiences to the end customer. By using cutting-edge technologies, utilizing cloud platforms, and working with innovative software developers,  automakers provide solutions to even the most demanding needs . And while our user experience is improving at an accelerated pace, there is still a broad list of challenges to tackle.

In this article, we dive into the technology behind the latest trends, take into account the most demanding areas of developing software in the cloud, and explain how proper solution empowers the change that affects us all.

Challenging determinants of the cloud revolution in automotive

Connecting with your car through a smartphone or utilizing information about traffic provided to your vehicle thanks to the platforms that accumulate data registered by other drivers is extremely useful.

Those innovative changes wouldn’t be possible without  cloud infrastructure . And as there is no way back from moving to the cloud, the transition creates challenges in various areas:  safety, security, responsiveness, integrity , and more.

Safety in the automotive sector

How to create a solution that doesn’t affect the safety of a driver? When developing new services, you cannot forget about the basics. Infotainment provided to vehicles is more advanced for every new release of a car and can be really engaging. The amount of delivered information combined with increasingly larger displays may lead to distraction and create dangerous situations. It’s worth mentioning that some of the colors may even impair the driver’s vision!

Integration with the cloud usually enables some of the remote commands. When implementing them, there are a lot of restrictions that need to be kept in mind. Some of them are obvious, such as you don’t want to disable the engine when a car is being driven 100km/h, but others may be much more complicated and unseen at first.

Providing security for car owners

Enabling services for your vehicle in the cloud, despite being extremely helpful to improve your experience, creates another way to break into your car. Everyone would like to open a car without using keys, but using a mobile phone, voice, or a fingerprint instead. And as these solutions seem modern and fancy, there is a big responsibility on the software side to do it securely.

Responsiveness enabling the seamless user experience

 Customer-facing services need to deliver a seamless experience to the end-user. The customer doesn’t want to wait a minute or even ten seconds for unlocking a car door. These services need to do it immediately or not at all, as an issue with opening the doors just because the system had a ‘lag’ is not acceptable behavior.

Data integrity is a must

Another very important concept associated with providing solutions utilizing cloud technologies is data integrity.  Information collected by your vehicle should be useful and up to date. You don’t want a situation when the mobile application says that the car has a range of 100km, but in the morning, it turns out that the tank is almost empty, and you need to refuel it before going to work.

How to integrate and utilize mobile devices to connect with your vehicle?

When discussing how to use mobile phones to control cars, a very important question occurs; how to communicate with the car? There is no simple answer, as it all depends on what model and version of a car it is, as depending on a provider, the vehicles are equipped with various technologies. Some of them are equipped with BLE, Wi-Fi Hotspots, or RFID tags, while others don’t offer a direct connection to the car, and the only way is to go through the backend side. Most of the manufacturers will expose some API over the Internet without providing a direct connection from mobile to the car. In such cases, usually, it’s a good practice to create your own backend which handles all API flaws. To do so, your system will need a platform to have a reliable solution.

When the limitation of hardware is met, there is always an option to equip the car with a custom device, which will expose a proper communication channel and will be integrated with the vehicle. To do so, it may use the OBD protocol. It gives us full control over the communication part, however, it’s expensive and hard to maintain the solution.

Building a platform to solve the challenges

There is no simple answer on how to solve the mentioned challenges and implement a resilient system that will deliver all necessary functionalities with the highest quality. However, it’s very important to remember that such a solution should be scalable and utilize cloud-native patterns. When designing a system for connected cars, the natural choice is to go with the microservice architecture. The implementation of the system is one thing, and partly this topic was covered in the     previous article   , but on the other hand, the very important aspect is a runtime, the platform. Choosing the wrong setup of virtual machines or having to deploy everything manually can lead to downtime of the system. Having a system that isn’t available for the customer constantly can damage your business.

Kubernetes to the rescue! As probably you know, Kubernetes is a container orchestration platform, which allows running workload in pods. The platform itself helped us to deliver many features faster and with ease to our clients. Nowadays, Kubernetes is so easily accessible that you can spin up a cluster in minutes using existing service providers like AWS or Azure. It allows you to increase the speed of delivery of new features, as they may be deployed immediately! What’s very important with Kubernetes, is its abstraction from infrastructure. The development team with expertise in Kubernetes is able to work on any cloud provider. Furthermore,     mission-critical systems can successfully implement Kubernetes   for their use cases as well.

Automotive cloud beyond car manufacturers

 Automotive cloud is not only a domain of car manufacturers. As mentioned earlier, they offer digital services to integrate with their cars, but numerous mobility service providers integrate with these APIs to implement their own use cases.

•  Live notifications
•  Online diagnostics
•  Fleet management
•  Vehicle recovery
•  Remote access
•  Car sharing
•  Car rental

The best practices of providing cloud-native software for the automotive industry

Working with  the leading auto motive brands and being engaged in numerous projects meant to deliver innovative applications. Our team have collected a group of helpful practices which make development easier and improve user experience. There are some must-have practices when it comes to delivering high-quality software, such as CI/CD, Agile, DevOps, etc., – they are crucial yet well-known for the experienced development team and we don’t focus on them in this article. Here we share tips dedicated for teams working with app delivery for automotive.

Containerize your vehicle

One of the things we’ve learned     collaborating with Porsche   is that vehicles are equipped with ECUs and installing software on them isn’t easy. However, Kubernetes helps to mitigate that challenge, as we can mock the target ECU by docker image with specialized operating systems and install software directly in it. That’s a good approach to create an integration environment that shortens the feedback loop and helps deliver software faster and better.

Asynchronous API

In the IoT ecosystem, you can’t rely too much on your connection with edge devices. There are a lot of connectivity challenges, for example, a weak cellular range. You can’t guarantee when your command to the car will be delivered and if the car will respond in milliseconds or even at all. One of the best patterns here is to provide the asynchronous API. It doesn’t matter on which layer you’re building your software if it’s a connector between vehicle and cloud or a system communicating with the vehicle’s API provider. Asynchronous API allows you to limit your resource consumption and avoid timeouts that leave systems in an unknown state.

Let’s take a very simple example of a mobile application for locking the car remotely.

 Synchronous API scenario

1.  A customer presses a button on the application to lock the car.
2.  The request is sent and is waiting for a response.
3.  The request needs to be delegated to the car which may take some time.
4.  The backend component crashes and starts without any knowledge about the previous request.
5.     The application gets a timeout.  
6.  What now? Is the car locked? What should be displayed to the end-user?

 Asynchronous API scenario

1.  The customer presses a button on the application to lock the car.
2.     The request is sent and ended immediately.  
3.  The request needs to be delegated to the car which may take some time.
4.  The backend component crashes and starts without any knowledge about the previous request.
5.  The car sends a request with the command result through the backend to the application.
6.     Application displays: “Car is locked.”  

With asynchronous API, there’s always a way to resend the response. With synchronous API, after you lose connection, the system doesn’t know where to resend response out of the box. As you may see, the asynchronous pattern handles this case perfectly.

Digital Twin

DigDigital Twin is a virtual model of a process, a product or a service, in case of automotive – a digital cockpit of a car. This pattern helps to ensure the integrity of data and simplify the development of new systems by its abstraction over the vehicle. The concept is based on the fact that it stores the actual state of the vehicle in the cloud and constantly updates it based on data sent from a car. Every feature requiring some property of vehicle should be integrated with Digital Twin to limit direct integrations with a car and improve the execution time of operations.

Implementation of Digital Twin may be tricky though, as it all depends on the vehicle manufacturer and API it provides. Sometimes it doesn’t expose enough properties or doesn’t provide real-time updates. In such cases, it’s even impossible to implement this pattern.

Software for Connected Cars - Summary

We believe that the future will look more futuristic than we could have ever imagined. Autonomous cars, smart cars, smart homes, every device tries to make our lives easier. It’s not known when and how these solutions will fully utilize Artificial Intelligence to make this experience even better. Everything connects as numerous IoT devices are connected which provides us with unlimited possibilities.

T  he automotive industry is currently transforming, and it isn’t only focusing on the driving experience anymore. There is a serious focus on connected mobility and other customer-oriented services to enhance our daily routines and habits. However, as software providers, we should keep in mind that automotive is a mature industry. The first connected car solutions were built years ago, and it’s challenging to integrate with them. These best practices should help focus on customer experience. Unreliable systems won’t encourage anyone to use it, and bad reviews can easily destroy a brilliant idea.

The automotive industry is experiencing a challenging transformation. We can notice these changes with every new model of a car and with every new service released. However, to keep up with the pace of the changing world, the industry needs modern technologies and reliable solutions, such as Kubernetes. And on top of that cloud-native application,     software created with the best practices by experienced engineers   who use the customer-first approach.