The News Wheel
No Comments

Microsoft and MIT Use AI to Improve Flaws in AV Technology

Decrease Font Size Increase Font Size Text Size Print This Page

As high-tech as self-driving vehicles are, they still have some weaknesses. For example, it doesn’t always know what to do in certain driving scenarios. That’s because there’s a knowledge gap between these vehicle’s AI-based training and real-life situations.
Microsoft and MIT have collaborated on a new model that can detect and help fill the gaps in AV technology. Whenever an automated vehicle (AV) doesn’t know how to behave in a situation, the AI part of the vehicle’s technology would compare what the vehicle was about to do with what a human driver would do. The tech would also adjust the vehicle’s behavior based on how closely it matched the human response. This, in turn, would improve the safe operation of the vehicle in challenging driving conditions.

Jumpstart Your Next Vehicle Purchase: Estimate your credit score here


The new model also incorporates real-time corrections. This simply means that a human driver could take over if, for some reason, the AV’s technology was confused or it started to respond in a dangerous way.

For example, the new model could help an AV know what the appropriate response would be for operating in snowy weather. It could learn from watching humans drive their vehicles slowly, and make turns and accelerate more slowly.
Though Microsoft and MIT are still perfecting the AI-based model, it promises to help bridge the gap between what responses an AV can determine and which ones they still need help determining. We anticipate more news in the days ahead as the partners fine-tune this advanced technology.

Upkeep Made Easy: Access your vehicle’s customized maintenance schedule


News Source: Engadget