Adaptive cruise control uses AI to learn driver’s behavior
Hyundai’s technology integrates artificial intelligence (AI) into the company’s Advanced Driver Assistance System (ADAS). The system, called Machine Learning based Smart Cruise Control (SCC-ML), will go into series production already in Hyundai’s next generation of vehicles, the company said in a press release.
SCC-ML combines AI and SCC into a system that analyzes and learns the driver’s driving behavior. Through this machine learning, the new technology independently regulates the speed according to the same pattern as the driver. The system works as follows: Sensors such as the front camera and radar constantly record driving information and send it to a central computer. This computer extracts the relevant details from the collected information in order to identify the driver’s typical driving patterns. This process of machine learning is controlled by AI. The system is programmed in such a way that it recognizes unsafe driving patterns and does not adopt them.
With currently available conventional cruise control systems, the driver manually adjusts the distance to the vehicle in front and the speed to be driven. Without machine learning, it is not yet possible to meticulously adapt the settings required for autonomous driving to the individual preferences of the driver. For drivers, this semi-autonomous driving feels strange and unfamiliar because it does not suit their driving style. This is why some drivers use the automatic speed control rather cautiously.
The algorithm captures three aspects of the manual driving pattern: distance to vehicles in front, how strongly the driver accelerates and how quickly he reacts to changes in the driving situation. External driving conditions and current speed requirements are also taken into account.
The system detects whether, for example, the driver is driving slowly in the city or in the overtaking lane of the motorway at a short or long distance from the vehicle in front. When analyzing the data, the system can differentiate between more than 10,000 different patterns and thus flexibly adapt to the driving style of each driver.
The system is so sensitive that it detects changes in the driver’s driving style and regularly adapts to them. In addition, SCC-ML is programmed in such a way that it does not learn any unsafe driving patterns and thus increases both reliability and safety.
With this new generation of cruise control systems in combination with the upcoming Highway Driving Assist System (HDAS), which supports automatic lane changes, Hyundai aims to achieve Level 2.5 on the road to fully autonomous driving.
Related articles:
Automotive industry outlook: Software passes hardware on the fast lane
Hyundai invests in Deep Learning computer vision startup
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
