The Lab

Velodyne VLP-32C lidar

The long-range lidar helps to detect far-away obstacles, e.g. cars and pedestrians. Velodyne VLP-32C lidar has 32 lidar beams and has detection range about 80 meters. Ground removal and clustering are applied to extract obstacles from the point cloud. Alternatively special neural network can be used to detect other vehicles and pedestrians.

Ouster OS1-128 lidar

The short-range lidar helps to detect close-by and small obstacles around the vehicle. Ouster OS1-128 produces very dense point cloud with its 128 rotating laser beams and has detection range around 40 meters. The precision of lidar measurements is 2-3 cm. Ground removal and clustering are applied to extract obstacles from the point cloud.

Allied Vision Mako G-319C cameras

The two cameras keep an eye on traffic lights. The left camera is pointed forward to see the farther traffic lights. The right camera is pointed slightly to the right to see the closest traffic lights next to the stopline. Traffic light locations are known from the high-definition map and are projected to the camera image. Then a patch with the traffic light is extracted from the camera image and neural network is used to classify if the light is red, green or yellow.

Aptiv ESR 2.5 radar

The radar uses reflections of radio waves to estimate the distance to the other vehicles ahead. In addition it also produces instant speed measurements of those vehicles using Doppler effect. Radar is very good in detecting big metallic objects, like cars, but struggles with humans and smaller objects. Additionally its distance measurements are very accurate in longitudinal direction, but lack precision in lateral direction.

NovAtel PwrPak7D GNSS device

The GNSS device determines the location of the car using satellite positioning and corrections from nearby base stations. The system makes use of Real-Time Kinematic (RTK) system with ESTPOS base station network to consistently achieve 5-10 cm positioning accuracy. The car needs to know its precise position on the high-definition map to follow the virtual rails drawn on the map.

AStuff Spectra computer

The computer runs the Autoware.AI software that makes all the driving decisions. Autoware.AI is open-source autonomous driving software stack that is built on top of Robot Operating System (ROS). Other than that the computer is just a normal PC running Ubuntu operating system. Over time we have made many tunings, bug fixes and adaptations to the Autoware.AI software.