- GM has filed a patent for autonomous driver-training technology, as first reported by Motor1.
- Instead of a human instructor, the system evaluates and trains novice drivers using various methods.
- GM’s system can give greater control to the driver-in-training if they achieve a high enough score.
Human driving instructors might soon be replaced by computerized ones. No, not full-blown robots, but rather autonomous vehicles. That’s the idea behind the driver-training technology that’s described in a patent that General Motors filed with the United States Patent and Trademark Office (USPTO), as first reported by Motor1.
GM’s patent describes autonomous technology that doesn’t require a living, breathing driving instructor. Instead, the system is designed to evaluate and train novice drivers by using a combination of sensors for the car’s autonomous features as well as monitoring how the driver-in-training operates controls such as the accelerator, brake, and steering. The system also uses a processor with a driving algorithm stored in the car’s memory. By comparing the trainee’s manual inputs with recommended driving instructions, the processor calculates a score.
From the sound of it, the score that GM’s autonomous driver-training feature calculates is similar to a grade given by a human instructor. The patent states that this score can be shared with third parties, too. Most importantly, it’s used to determine if the driver meets predetermined thresholds. It then selectively gives the driver greater control of various vehicle functions based on how their score compares with the preset thresholds. That sounds a lot more intuitive than the trial-by-fire method of letting young Kyle or Kelly learn how to navigate busy intersections or merge onto the freeway before they’re fully prepared.
While we’re not entirely sure if this system will be offered on personal vehicles, the patent points out that the technology could be used as part of a fleet for driver-education purposes. It also mentions the possible advantages of using autonomy to teach new drivers versus human instructors. Not only can real people be prone to bias, the patent suggests, but traditional training techniques might be more expensive and less efficient than autonomous techniques.
We also don’t know if GM has any kind of timeline for rolling this technology out to the public, but it is encouraging to see that in the future the technology can be used to train people how to drive rather than mainly taking that ability away from them.
This content is imported from {embed-name}. You may be able to find the same content in another format, or you may be able to find more information, at their web site.
This content is created and maintained by a third party, and imported onto this page to help users provide their email addresses. You may be able to find more information about this and similar content at piano.io
Source: Motor - aranddriver.com
