Our Autonomous Valet Parking (AVP) team attended the Cenex-Connected Automated Mobility (Cenex-CAM) event on 4-5 September 2019, to demonstrate the progress of our 30-month Autonomous Valet Parking system project.
The project seeks to navigate a self-driving car to an open parking space, autonomously park itself and respond to driver commands. The project is led by Dr. Brian Holt, Head of Indoor Mapping at Parkopedia. Brian leads the Research and Development team that are developing detailed indoor maps for covered car parks to enable Autonomous Valet Parking to become a reality to the general public in the future. The project is funded by InnovateUK and the Centre for Connected and Autonomous Vehicles and is due to end on Saturday 31 October 2020.
At the event, we showed how the autoware-based software on the AVP Consortium Streetdrone was able to control the vehicle by following waypoints consistently and accurately. The demonstration consisted of three parts and reflected how we believe AVP will be implemented in the future.
- The vehicle following a predefined set of waypoints to the designated parking space, having been dropped off at a designated drop-off zone by its driver.
- The vehicle exiting the parking space and driving to the pick-up zone (where the vehicle’s regular driver would collect it).
- A test of our automatic emergency braking, using the front-centre ultrasonic sensors on the vehicle.
The public demonstration was an important milestone for us to show our ability to control the vehicle using a PID controller for longitudinal (throttle) control and pure-pursuit for lateral (steering) control.
We have achieved a great deal since the project began and with thirteen months left, we aim to work on the navigation and localisation using maps. We are due to give a final demonstration of the end solution in Autumn 2020.
Read more about the demonstration and AVP project here.