Carmakers have adopted interactive artificial intelligence (AI) services such as infotainment systems embedded with AI voice assistants to help drivers control radio, air conditioning and navigation verbally. However, such smart AI features can interrupt driving or distract the driver by trying to engage in a conversation regardless of road conditions ahead.
The Korea Advanced Institute of Science & Technology (KAIST) said in a statement on Monday that its research team led by industrial & systems engineering professor Lee Ui-chin has developed an interactive AI system that collects data generated by sensors such as the movement of the steering wheel, the operation of brake pedals and environmental factors.
"It is a technology which predicts the right time to engage in an interaction based on basic in-vehicle sensor data," Lee said, adding the new technology can be applied to in-vehicle infotainment systems. The new system provides simple information such as the weather in low traffic situations and tries to interact with the driver without jeopardizing the vehicle's safety.