Self Driving Car with Jetson AGX Xavier... what is stopping it?

I was reading this forbes article:
https://www.forbes.com/sites/patrickmoorhead/2018/12/17/robotics-and-smart-cities-just-got-a-lot-more-real-with-nvidia-jetson-xavier-edge-compute-module/#46f98b037718

One excerpt from the article:
"While Jetson AGX won’t be in self-driving cars, the point was to demonstrate the dramatic improvement in performance of Jetson AGX Xavier. "

What is it about the Jetson AGX Xavier that makes it not possible for use in Self-Driving Cars or Medical Imagery? Arn’t there vision demos available to the AGX Xavier? Is it a performance issue that makes Jetson different from DRIVE AGX? Or something else?

Hi,

Jetson is designed for embedded use-case.
It’s more suitable for Robotics or IoT field.

We have another platform called DRIVE AGX Xavier.
This platform is designed for self-driving car:

Thanks.

Given the specifications, it looks like DRIVE AGX Xavier (30 TOPS) has less TOPs than the Jetson (32 TOPS).

I see this:
https://www.nvidia.com/en-us/self-driving-cars/drive-platform/hardware/

There are different models of the drive variants, but I see 320 TOPS for the Drive Pegasus AGX, but the same 30 TOPS you mention for the “ordinary” drive AGX Xavier. I will speculate that the temperature range and other details were “ruggedized” somewhat for this, and as a result clocks might be lower.