Unmanned Systems Technology 013 | AutonomouStuff Lincoln MKZ | AI systems | Unmanned Underwater Vehicles | Cosworth AG2 UAV twin | AceCore Neo | Maintenance | IDEX 2017 Show report

22 Dossier | AutonomouStuff Lincoln MKZ The offset uses a precise measurement in reference to the centre of the rear axle, and the coordinate transformation happens in the middleware. That is essential to the modular approach, as it allows the mounting of the sensors to be moved easily. The mounts are 3D printed to fit in different locations. Other laser sensors The AS MKZ also uses rotating laser sensors from Ibeo Automotive Systems for different functions. “These have always been one of my favourites,” says Hambrick. “You get a directionally focused beam with four layers and a 110 º FOV, so it’s a smaller amount of data than the Velodyne. We use them on the bumpers to cover the blind spot of the Velodyne on the roof.” These sensors have a narrower FOV than the Velodyne and a longer range, of 200 m, as well as integrated object tracking. Hambrick says, “We fuse the data from six of those around the car, and use that with the object tracking to fuse the data into one scan.” The raw data is fed from the six sensors – three on the front bumper and three on the back – to an Ethernet switch and then to an ECU from Ibeo that acts as a fusion engine. This combines the raw data from the sensors and identifies objects in the point cloud, the output being both the objects and the raw data. This provides detection and tracking of other road users with detailed information on the position, motion and shape of surrounding road users as well as static background objects. It detects cars, trucks, bikes and pedestrians, and provides estimated uncertainties for each object. April/May 2017 | Unmanned Systems Technology Sensors are installed in the MKZ using 3D-printed mounts The front and rear bumpers of the MKZ are fitted with Ibeo laser scanners