Reservation Enhanced Autonomous Valet Parking Concerning Practicality Issues

Advances in automotive industry as well computing technology are making autonomous vehicle (AV) an appealing means of transportation. Vehicles beyond the traditional source commute, and leveled up to smart devices with capability. As one compelling features AVs, valet parking (AVP) allows for navigating car automatically without human interventions. Within this realm, long-range AVP (LAVP) extends auto-parking a much larger scale compared its short-range counterparts. It is worth noting that AV mobility pivotal concern LAVP, involving dynamic patterns related spatial-temporal features, such varied drop-off (or pick-up) demands diverse customer journey planning. We herein target critical decision-making on where park drop/pick-up upon requirements during their journeys. Concerning practice parks equipped limited space, we thus introduce reservations enable accurate estimations future states. An efficient LAVP service framework enhanced then proposed. Benefited from intelligent predictions, load can be accurately predicted greatly alleviated at individual parks, thereby avoiding overcrowding effectively. Results show significant performance gains achieved under proposed scheme by comparing other benchmarks, respect reduced waiting duration available experiences terms traveling period, etc. In particular, number parked vehicles across network effectively balanced.