AI and the automotive industry
According to the IBM website, artificial intelligence is “technology that allows computers to simulate human intelligence and human problem-solving abilities. On its own or combined with other technologies (e.g., sensors, geolocation, robotics), AI can perform tasks that would otherwise require human intelligence or intervention.
Taken to the field of the automotive industry, the application of AI has autonomous driving as its main objective. “The main objective of AI is not the integration of a chatbot like ChatGPT, but rather it is a system that integrates information from various sensors such as cameras, LIDAR sensors, Radar, etc., to make predictive decisions about what the user should do. vehicle. “It is used in autonomous systems so that a car, for example, can take you from point A to point B without human intervention.”he explained Arturo Deza, professor at the Faculty of Computing at the University of Engineering and Technology (UTEC).
/
On the other hand, the expert pointed out that AI is not only found in autonomous vehicles, but also in non-autonomous ones. “There are also applications in non-autonomous vehicles, such as assisted navigation. When we use Google Maps or Waze, we are using AI to optimize transportation routes.”he declared.
Another aspect of AI in transportation is the use of sensors that assist in driving. This can be useful in unforeseen situations, where the driver must make quick decisions and understand when exceptions need to be made. This occurs, for example, when there is a lot of traffic, drivers must ignore traffic signs, giving preference to the manual signals of a police officer (and they must run the red light on purpose), adapting their behavior depending on the circumstances.
AI must, in certain cases, recognize these patterns and understand, from a cognitive perspective, when a traffic rule needs to be adapted to maintain harmony in the flow of vehicles. Those are some of the uses that AI is having in driving and the transportation industry, especially in the context of autonomous vehicles and navigation systems.
Advantages of AI
One of the main advantages is that with AI many vehicles are beginning to implement intelligent driving. You have to do less and less to obtain a quick response from the browser, always with the aim of not losing concentration while driving. A clear example is found in voice assistants, which allow you to control multiple vehicle functions such as climate control, window and seat control, among others. Although, it also allows you to make calls or search for a specific place.
Modern vehicles include advanced driver assistance system (ADAS) packages, which are passive and active safety systems that are designed to eliminate the human error component in the operation of vehicles of many types. These mechanisms use advanced technologies to assist the driver while driving in order to improve their performance.
/
To do this, they use a combination of sensors that allow them to perceive the world around the car and then provide information to the person behind the wheel or take action if necessary.
Among the ADAS applications essential for safety included:
- Pedestrian detection or avoidance
- Traffic sign recognition
- Lane departure warning or correction
- Automatic emergency braking
- Blind spot detection
According to Kia’s website, these ADAS have been incorporated into the latest interface standards to run multiple vision-based algorithms that support real-time sensor fusion, vision co-processing and multimedia subsystems.
AI also provides certain advantages in the manufacturing of cars, since it is beginning to be used to design them in a more efficient way. “AI is being used to optimize the construction of the parts, making it more aerodynamic or resistant to certain thermal phenomena”said the expert.
Likewise, AI is being applied to the entire supply chain, from identifying where in the world the materials such as minerals and semiconductor metals needed to make microchips come from, to autonomous driving and assisted driving systems in cars.
Disadvantages of AI
A disadvantage is that most of these systems learn from how humans drive, without explicit understanding. For example, a car should recognize the presence of a pedestrian or a street vendor and act accordingly. However, these systems sometimes do not know how to react in atypical situations and can make wrong decisions.
“This occurs because the perception system has not been able to correctly understand a scene and therefore an accident occurs; or in a very complementary way, when the system does perfectly recognize everything that happens in the scene, but makes a wrong decision that was not previously mapped in its database, given the complexity of the emerging situation.”detailed the UTEC expert.
Additionally, Deza added that another drawback is that an autonomous vehicle (powered by AI) could easily cost US$300,000, since it requires a series of elements and systems for its operation.
Adaptability in Peru
The Peruvian customer is already accustomed to using non-autonomous vehicles that have a certain degree of artificial intelligence (AI), such as when we use Google Maps or Waze. However, how the Peruvian customer would react to a completely autonomous car is still unknown, because this technology is not available locally.
Although there are already Teslas in Peru, although very few, the Full Self-Driving (FSD) system does not work properly due to the characteristics of the traffic and driving style in the country, which presents a challenge for its implementation. In that sense, Deza explained that there is currently a collaboration between UTEC and Artificio to solve this problem.
“We are using the UTEC supercomputer with Artificio data to see how different artificial intelligence models in the global world react to the traffic of Lima, Cusco and Cajamarca”he mentioned.
AI and car brands
The vast majority of car brands already implement AI in the manufacturing line of their models or in the features of their cars that they offer to their users. In that sense, the progress that the Volkswagen group and the Daimler group are making with Mercedes-Benz are notable.
In the case of Volkswagenthe German giant created its own artificial intelligence company called “AI Lab” in early 2024. According to its website, it aims to devise new products related to AI and collaborate with technology companies when the need arises.
Similarly, VW also announced at the beginning of this year that it will begin to integrate the ChatGPT AI system in its electric vehicles (ID.7, ID.4, ID.5 and ID.3) and in the new Tiguan, Passat and Golf. All of these will have the voice assistant called IDA, which allows you to control the infotainment, navigation and air conditioning systems, and can also answer general questions.
/
AI not only powers navigation or secondary features, but also security. The VW Teramont integrates emergency assistance that acts if the driver does not respond to the signals that the vehicle emits when it is moving. For example, if the driver faints, the assistant will act and take control of the vehicle until it comes to a controlled stop off the road.
On the other hand, Mercedes-Benz It has also developed security assistants based on AI, although one of its greatest achievements in this regard is the creation of its system MBUX (Mercedes-Benz user experience). According to the German firm, this system uses advanced software and generative artificial intelligence to create an even more natural and intuitive relationship between the person and the vehicle. All with the aim of making our lives easier.
The MBUX virtual assistant, which is also powered by AI, became popular because to activate it the driver had to say “Hey Mercedes” to activate some functions, although now it is enough to simply speak for the assistant to help you. “This is a predictive system that can proactively offer responses and suggestions, based on situational context, learned behavior and the use of generative AI,” reads the Mercedes-Benz website.
/
In addition, MBUX finds support in 3D graphics for more intuitive communication, as well as Surround Navigation to offer an augmented reality view of the vehicle’s environment in real time with powerful 3D graphics created with the Unity video game engine. . For example, in traffic situations.
In the same way, autonomous driving is part of the present and future of Mercedes-Benz. “Currently, we are developing automated driving capabilities through the use of algorithms that analyze real-time data from multiple sensors and cameras, allowing the vehicle to make informed decisions”said a spokesperson for Mercedes-Benz Peru.
Soon we will talk to our car
As mentioned, AI offers multiple benefits to the automotive industry. It allows you to reduce production costs, optimize vehicle operation, integrate safety alerts such as ADAS, improve the user interaction experience and take another step towards autonomous driving.
Advances in generative AI will allow us to maintain a natural conversation with our vehicle for various purposes in the near future. Currently, some cars already have voice assistants.