A night drive with Tesla driver monitor system watching. The text cutout on the top left is Tesla DMS probabilities output showing what the system is thinking really happening in the frame. The probabilities are recomputed about 15 times per second so it does jump quite a bit.

KENOSHA, Wis. —Kenosha County Sheriff's deputies pulled over a Tesla driver who fell asleep while on autopilot.

Investigators said they spotted the driver on I-94 north near State Highway 158 just before 8 a.m. Sunday.

Deputies said the Illinois man was asleep behind the wheel and woke up once the car was pulled over.

Deputy David Gomez told the driver he followed him for two miles and saw him slumped over in the driver's seat.

"I looked over and noticed his head was turned away from me and looking down," Gomez said.

He said he clocked the Tesla at 82 mph.

A 911 caller reported the car would speed up when not around other cars and then slow down again as it approached traffic.

Another 911 caller said she saw the driver with his head back and mouth open.

The deputy reported the driver finally noticed he was being stopped when the deputy pulled alongside the Tesla for the second time.

The Tesla driver was identified as Mitul Patel, 38, of Palatine.

He was questioned and initially denied falling asleep while the car was driving itself.

Gomez said Patel did not show signs of impairment but stated he was tired.

Patel told Gomez he was on his way to work as a FedEx driver in Cudahy.

He repeatedly denied sleeping and tried to explain himself.

"If someone's not holding the steering wheel for 30 seconds it will automatically turn off autopilot," he told Gomez.

Tesla mandates that the operator keep their hands on the steering wheel at all times and always maintain control of the vehicle while utilizing the autopilot feature.

"I get the auto option, but you still have to be awake and doing things," Gomez said.

The driver was cited $187.90 for inattentive driving and released.

Gomez said this was not the first time Patel was accused of driving while asleep and on autopilot.

"I looked at the history and the vehicle with this registration was attempted to be pulled over two different times in February and August," Gomez said. "Both of them reported the driver appeared to be head down, sleeping."

Patel's car was towed because deputies said they felt it was unsafe for him to continue driving.

When reached by WISN 12, Patel said he did not have any comment on the story.

​​​​​​​Tesla today announced the official transition to “Tesla Vision” without radar on Model 3 and Model Y.

In the process, the automaker warns of some limitations on Autopilot features at first.

Over the last few months, Tesla has been talking about transitioning its Autopilot and Full Self-Driving technology to solely using computer vision based on cameras and not relying on its front-facing radar anymore.

Now Tesla announced that they are removing the radar altogether on new Model 3 and Model Y vehicles produced in the US:
The automaker warned that the transition will result in some basic limitations of Autosteer, an Autopilot feature, and Smart Summon, a Full Self-Driving package feature:

• Autosteer will be limited to a maximum speed of 75 mph and a longer minimum following distance.
• Smart Summon (if equipped) and Emergency Lane Departure Avoidance may be disabled at delivery.

As usual, Tesla plans to improve on the technology through over-the-air software updates in order to restore those features and go beyond current capability:
The latest iteration of Tesla Vision was first developed as part of Tesla’s Full Self-Driving Beta program that was tested in its early access program.

CEO Elon Musk has been hyping the vision-only update as “mind-blowing.” He insists that it will lead to a true level 5 autonomous driving system by the end of the year, but he has gotten that timeline wrong before.

Tesla also confirmed that Model S and Model X, as well as Model 3 and Model Y built in China, will also transition to Tesla Vision without confirming the timing.

As we previously reported, the idea of moving to only cameras using computer vision is that the only known system that can drive right now is the human brain. It relies on input from human eyes, which are closer to cameras than anything else.

With cameras all around the vehicle with different fields of view, Tesla can achieve greater vision than humans, and the problem becomes only solving computer vision, which the automaker believes it is on the way to solving.