The AVP Project ended on 31 October 2020. Please visit Parkopedia’s Indoor Maps page for further updates.
As autonomous cars start to become a reality, one of the unanswered questions remains – where and how will those cars park?
This consortium’s “Autonomous Valet Parking” project seeks to develop Highly Autonomous Driving maps to support indoor navigation and localisation.
Autonomous Valet Parking (AVP) is functionality which allows a driver to be dropped off in a multi-storey car park or their final destination and the vehicle to then park itself autonomously.
Estimating the vehicle’s current position is more difficult in multi-storey carparks where GPS signals cannot be received, which means that the vehicle must rely on other sensors and localisation based on visual objects and features present in maps. This is an open problem in the automotive industry which must be solved to enable SAE Level 4 AVP deployment.
This consortium’s key objective was to identify obstacles to full deployment of AVP through the development of a technology demonstrator. We achieved this this goal by:
- Developing automotive-grade indoor parking maps required for autonomous vehicles to localise and navigate within a multi-storey car park.
- Developing the associated localisation algorithms – targeting a minimal sensor set of cameras, ultrasonic sensors and inertial measurement units – that make best use of these maps.
- Demonstrating this self-parking technology in a variety of car parks.
- Developing the safety case and prepare for in-car-park trials.
- Engaging with stakeholders to evaluate perceptions around AVP technology.