News

SZE’s drone research: Zalazone Smart City survey and first video streams

A next stage of forerunner drone development at Széchenyi István University was the survey of ZalaZONE Smart City (targeted to be the place of final demonstration) and the acquisition of the first aerial videos
In the project “Development of innovative automotive testing and inspection competences in the West-Hungarian region based on the infrastructure of the Zalaegerszeg Automotive Test Track GINOP-2.3. 4-15-2020-00009”, in frame of the Széchenyi István University part of the FT2 “Autonomous Near-Earth Aerial Solutions” subproject – in cooperation with Institute for Computer Science and Control in Eötvös Loránd Research Network –the survey of Zalazone Smart City, practicing of test situations and acquisition of aerial videos were performed after the previous forerunner developments.
The Smart City is targeted to be the place of final demonstration.
The forerunner drone means a camera equipped autonomous aerial vehicle flying in front of an emergency ground vehicle and observing the traffic ahead. The main goal is to notify the driver about hidden dangers such as non-braking vehicles behind the building walls in an intersection.
Smart City survey means the exploration of the scene, the design of test situations, the mapping of the streets with high precision GPS and the measurement of lane widths. Based-on these measurements the route of the emergency ground vehicle can be designed in advance and the waypoints (positions of intersections) can be sent to the drone.
Smart City provides only one intersection with building coverage from all sides so the designed test situations are fitted to this intersection. Two situations are planned to be tested in four different modes.

The two situations are:
A. The emergency vehicle approaches from the STOP sign (red car in the video) then
turns left. Another vehicle (silver car in the video) comes from its right and has to
give the right of way.
B. The emergency vehicle approaches from the STOP sign (red car in the video) then turns left.

Another vehicle (silver car in the video) comes head on then turns right
and has to give the right of way. In the second case we reversed the driving direction of the one way street which is feasible on a closed test ground.
The four modes are:

  1. The endangering vehicle approaches slowly and stops in time.
  2. The endangering vehicle approaches and stops in a short time.
  3. The endangering vehicle approaches and stops with emergency brake.
  4. The endangering vehicle approaches and does not stop, the emergency vehicle should stop.
    These four modes are recorded to determine which situation can be handled based-on the camera images and which not. Recordings were done hovering above the intersection.

Based-on the video streams the tuning and testing of the decision system can be done. The video shows the A / 1-4 and B / 2 and 4 cases.