Unmanned Systems Technology 003 | UAV Solutions Talon 120 | Cable harnesses | Austro Engine AE50R and AE300 | Autonomous mining | AUVSI 2015 show report | Transponders | Space systems

58 the system slows the vehicle to a safe stopping point when an obstacle is detected and await operator intervention, but Cortex dynamically plans the safest and most efficient pathway around an obstacle, as identified from sensor data, so that no operator approval is necessary. This level of control is achieved using 3D terrain maps that are generated by combining, or ‘fusing’, GPS data and satellite images with raw data from the stereo cameras, Lidar laser rangefinders and radar on the vehicle. The Lidar system, called Forecast, was developed to identify potential obstacles and integrate seamlessly with Vantage to map the 3D space surrounding a vehicle. Software in the Forecast unit then creates a map and can perform rudimentary obstacle detection tasks. This data fusion allows the vehicle to navigate accurately, both outdoors and in mining tunnels. All this data feeds into a system of robotic components called NAV. As most modern haul trucks have electronic networks internally, NAV can use the existing CAN bus system to direct vehicle functions, so it is able to control the drive transmission, steering, acceleration, braking and dump bed, and monitor the real-time health of the vehicle. There are several other elements to the ASI approach. As with the black box in aircraft, ASI’s BlackBox software (which comes as an upgrade) enables Cortex to record critical data logs to a MicroSD card for playback and analysis. There is also a protective casing called SpotPoint that mounts on the roof of the vehicle to house the autonomous components such as the GPS and radio antennas, cameras and warning beacons. Much of this technology evolved from a research project with the US Army and the GPS and Vehicle Dynamics Laboratory at Auburn University, Alabama, which specialises in vehicle and non- linear dynamics. Its Stability Control Improvement and State Detection for Autonomous Vehicles project combined a dynamic vehicle model with sensor data to enable a robotically controlled vehicle to avoid obstacles, detect and counter yaw instability, and avoid vehicle rollovers. Then there are the vehicle’s mechanical elements to consider. ASI provides a steering ring for mechanically operated vehicles that also allows robotic steering. There is also an onboard as well as remote emergency stop, which will immediately halt all vehicle operations or allow an operator to switch between manual and automatic control to take over control if needed.  Summer 2015 | Unmanned Systems Technology ASI’s NAV module brings all the autonomous vehicle control elements together – including the SpotPoint system on the roof – in a rugged box (Courtesy of ASI) The management of autonomous vehicles in software is a key element of the overall control system (Courtesy of ASI) 3D terrain maps are generated by combining, or ‘fusing’, GPS data and satellite images with data from cameras, Lidar and radar