One of the company’s autonomous vehicles, which was being tested near its headquarters in Mountain View, CA, struck a city bus on a public street, according to a newly released accident report. (The accident took place on Valentine’s Day.)

This is the second time this year a Google self-driving vehicle has been in a wreck – and there have been more than a dozen accidents with the cars since the self-driving car project began. Google has said all of those were caused by humans. (However, human drivers have prevented the autonomous vehicles from causing an accident 13 times so far.)

While the current report does not assign blame to either vehicle, Google (GOOG) said the car, a Lexus SUV equipped with Google’s sensors (called a Google AV), was attempting to navigate around sandbags in the street and was pulling into another lane when it struck the bus on the right side. A test driver was in the car, but expected the bus to yield.

The self-driving car was going 2 mph and the bus was traveling 15mph. There were no reported injuries in the fender-bender, but the Google AV’s front left quarter panel got banged up.

The local Department of Motor Vehicles is investigating the accident to determine who is to blame. If Google’s vehicle is cited, it will be the first time an autonomous vehicle has been found responsible for a collision.

To date, the self-driving car project has driven over 1 million miles in its three test cities – Mountain View, Austin, TX and Kirkland, WA. (The company is planning to add four more test cities soon.) Google says its cars are equipped with “sensors designed to detect objects as far as two football fields away in all directions, including pedestrians, cyclists and vehicles”.

It hopes to introduce the cars to consumers by 2020 – and seems to be making progress. The National Highway Transportation Safety Administration has ruled that the artificial intelligence guiding the car can be considered a legal driver under federal law.