From Random
As I said this is way beyond me as the cars I used to work on all the time only sensors were the driver.
From the looks of the manual there are at least 6 different sensors involved with the steering. Each of those sensors have between 2 and 6 possible failure modes. Add all the wiring, the control module and actual moving parts into the mix and an intermittent failure is almost impossible to identify through guess work.
To go back to what has been said I do not think you can diagnose this without capturing the failure codes with Techstream. There is just so much going on.
For example
"DESCRIPTION
The power steering ECU assembly receives vehicle speed signals from the skid control ECU via CAN communication. The power steering ECU assembly provides appropriate assisting force in accordance with the vehicle speed, based on the signals."
Your problem could be anyplace between a wheel speed sensor and the steering column. Even a loose or slightly corroded wire could be the issue. The only way to narrow it down would be to get the codes.
There was a big recall for the steering in Australia and all that was happening was that the position of some wires could cause small amounts of debris to create interference in the signal and the steering would cut out. No this does not apply to your car but that is how little it takes for the system to screw up.
https://www.toyota.com.au/news/toyot...ted%20vehicles.
What I am confident in saying is that in all probability the motor is not the issue.
Now that you have everybody's attention we should probably take this discussion back to your failure tech thread so that if an answer is found it is not lost to posterity in the wastelands that are random thoughts.