Scientists note that humans are still wary of handing over control entirely to machines. Voice assistant will increase confidence in autonomous transportation by allowing passengers to direct the vehicle.
The challenge is to teach the system to understand natural speech and relate it to the situation on the road. The researchers used the Talk2Car dataset, where you have to point to road sections according to text instructions.
The CAVG model was created to solve this problem. It uses computer vision to highlight areas in a camera image and then analyzes how they match commands. Particular attention is paid to understanding the emotional context of teams and difficult situations on the road. The model uses modern language technologies such as GPT-4V to interpret commands correctly.
To test its effectiveness, the model was tested in difficult conditions: at night, in heavy city traffic, with unclear commands and poor visibility.
CAVG showed strong results, outperforming existing solutions. They plan to improve the research by adding more types of data to the system, such as a top view (bird’s eye view) and information on the trajectory of movement. This will increase control accuracy and make autonomous cars even more reliable.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
