As part of the government-backed ‘Autonomous Valet Parking’ project, Parkopedia is leveraging the open-source self-driving stack, Autoware, to validate the indoor and outdoor mapping required for complex automated parking.
At LCV, the StreetDrone Twizy performed several manoeuvres using Parkopedia’s validated versions of the Autoware distribution including: Following pre-defined waypoints to a parking spot, exiting a parking spot to to meet a passenger and finally, emergency braking upon detecting a pedestrian within the intended path.
These demonstrations are a fundamental leap forward for the project, and represent an important milestone in vehicle control using a PID controller for longitudinal (throttle) control and pure-pursuit for lateral (steering) control.
The remaining 13 months will be spent concentrating on localisation and navigation using maps, with a final demonstration planned for Autumn 2020, however this latest development represents a massive step along the way towards a true autonomous parking experience.
StreetDrone is looking forward to continuing to support Parkopedia in their upcoming developments with their StreetDrone Twizy platform. Fore information on the StreetDrone Twizy, please do get in touch.