Visteon Corporation has developed a cockpit concept that incorporates Artificial Intelligence (AI) to deliver an enhanced driving experience. Visteon’s Human Bayesian Intelligence Technology (HABIT) system employs machine learning algorithms that are cognizant of the specific driver and the surrounding environment.
The HABIT system continually learns as it processes a driver’s selections of climate temperatures, radio stations, phone call tendencies and other unique behaviors depending on the outside temperature and time of day. It factors in the individual’s historical inputs to present a human-machine interaction (HMI) that is customized for the driver. The system also learns the driver’s tastes — even when he or she is not in the vehicle. For example, HABIT registers activity like music that the driver has listened to using his or her on-line music library or Internet radio.[image_frame style=”framed_shadow” align=”center” alt=”HABIT Cockpit Concept” title=”HABIT Cockpit Concept” height=”250″ width=”600″]https://www.car-engineer.com/wp-content/uploads/2013/05/HABIT-Cockpit-Concept.jpg[/image_frame]
“The goal of HABIT is to become an experience that improves each time the driver uses the ever-aware system,” said Shadi Mere, innovation manager at Visteon. “With vehicle manufacturers striving to deliver a more personalized driving experience, the HABIT cockpit concept demonstrates how your car can learn and grow with you over its lifetime.”
During a recent research clinic, more than 70 percent of survey respondents had a positive initial reaction to the HABIT concept. Respondents liked the anticipatory learning of the system and the natural voice interaction, combined with voice shortcuts, which aligned with their expectations of voice commands similar to the ones on their smartphones and other devices. Visteon’s cockpit concept incorporates high-end graphics and animation designed to improve interaction with mobile devices.
[youtube url=”http://youtu.be/cq1n4D0AHBU” fs=”1″ rel=”0″ showinfo=”0″]
Source: Visteon
[titled_box title=”Romain’s opinion:”]
This sounds great but it has to be robust enough as it should not annoy at all the driver by wrong choices or set-up. By the way, I’m wondering what kind of self-learning algorithm is used in this project? Neural Network? Pontryagin Minimum Principle?[/titled_box]
