Enlarge /. Tesla Model S drives on the freeway in Silicon Valley.
A Massachusetts man is negligent after his Tesla crashes into a police car parked on the side of the road. According to a state trooper, the man turned on Tesla's autopilot technology and said he "didn't have to be careful". The crash occurred in December, but the accused Nicholas Ciarlone was recently charged with the incident.
NBC Channel 10 in Boston reports that the police car was parked on the left side of Route 24, a shared freeway in West Bridgewater, a city about an hour south of Boston. The State Trooper had just run over student Maria Smith and asked for her registration documents when the Tesla slammed into his off-road vehicle.
This caused an accumulation, with the police car crashing into the student's vehicle. The soldier was pushed back against the roadside concrete barrier, but was not seriously injured. Smith said she had glass in her hair when the back window broke. And Smith told NBC 10 that the officer could have been slightly more seriously injured.
"If my car had continued to push forward, it would probably have been crushed by it," she said to NBC 10 Ryan Kath.
Teslas repeatedly bumps into stationary vehicles
This is a known issue with Tesla's autopilot – and with similar products from other automakers. In 2018, there were at least two incidents where autopilot crashed into parked fire engines. Fortunately, none of the crashes resulted in death. In 2019, a Tesla Model 3 crashed into a parked police car on a Connecticut freeway. Police said the driver "checked his dog in the back seat" before the accident.
In two other cases – one in 2016, the other in 2019 – Teslas crashed into the sides of tractor units that crossed in front of the vehicles. In both cases, the Teslas slipped under the trailer, sheared off the tops of the vehicles and killed the drivers immediately.
As I wrote in 2018, this is not unusual behavior for adaptive cruise control systems. These systems often work by adjusting the speed of vehicles in front. This is fairly easy with radar that can measure the speed of another vehicle directly. Such systems can completely ignore stationary vehicles because radar has poor angular resolution and cannot distinguish stationary objects near the road (such as a concrete lane divider) from an obstacle in the vehicle's lane. This works well enough most of the time, but in rare cases can lead to a car being parked in the lane and the driver not paying attention.
Tesla is committed to developing a more sophisticated self-driving system that fully understands the environment. Hopefully, Tesla's "fully self-driving" software will eventually recognize such a situation and respond appropriately. But at least last December, the technology still seems to be in the works.