How LIDAR Can Reduce Crashes of Driverless Vehicles


Photo Credit via Flickr


Tesla’s Model S crashed into the back of a stopped firetruck on Monday at the 405 freeway in Los Angeles. The driver told the authorities that the car was in autopilot mode at that time. The incident led the United States National Transportation Safety Board to conduct its own investigation. With the recent event, are autopilot systems truly safe to take on the road?

Cars with autopilot systems have a manual that warns drivers about certain conditions when activating autonomy. According to the owner’s manual of Model S, the car is not designed to handle this exact scenario: “Traffic-Aware Cruise Control cannot detect all objects and may not brake/decelerate for stationary vehicles, especially in situations when you are driving over 50 mph (80 km/h) and a vehicle you are following moves out of your driving path and a stationary vehicle or object is in front of you instead.”

It is the same with the Volvo’s Pilot Assist system. According to the official website of Pilot Assist, the system cannot handle all traffic, weather, and road conditions. Moreover, the driver should always be aware of traffic and intervene if the system fails to maintain the necessary speed and distance. Apparently, the issue may not be from autopilot systems, rather, in detection and ranging tools of the vehicle.

“The radars they use are apparently meant for detecting moving objects (as typically used in adaptive cruise control systems), and seem to be not very good in detecting stationary objects," said Raj Rajkumar, a researcher for autonomous driving at Cargenie Mellon University.

This is where LIDAR or light detection and ranging comes in, equipped with several sensors and lasers for precise mapping. It is a necessary tool for collecting data about traffic, road and weather conditions, and obstacles around the vehicle. With LIDAR, a vehicle can benefit from the following features:

- No geometrical distortions compared to a side-looking radar.

- Capability to collect elevation data in a dense forest.

- Quick and accurate collection of data.

- Uses an active illumination sensor.

- Can be integrated with other data sources.