Adaptive Occupant Experience and Concurrency Management in Fully Autonomous Robotaxi Services
Author(s): Ronak Indrasinh Kosamia
Publication #: 2503059
Date of Publication: 08.09.2021
Country: USA
Pages: 1-31
Published In: Volume 7 Issue 5 September-2021
DOI: https://doi.org/10.5281/zenodo.15086900
Abstract
Recent progress in fully autonomous vehicles—often referred to as robotaxis—has spurred the deployment of ride-hailing services without human drivers. This shift, while revolutionizing mobility, raises new questions about occupant experience, concurrency in multi-passenger scenarios, and ensuring safety in the absence of a dedicated driver. This paper presents a framework for adaptive occupant experience and concurrency management, emphasizing occupant detection, seat assignment, resource allocation, and automated conflict resolution. By merging onboard inference with microservice-based aggregator modules, our approach handles occupant classification, occupant-based personalization, and environment-aware route planning in real time. We adopt ephemeral occupant data storage to safeguard privacy, discarding raw sensor inputs immediately after local inference. Preliminary evaluations suggest that occupant-based concurrency can reduce disputes and enhance passenger comfort in multi-rider robotaxis, while maintaining minimal overhead on embedded hardware. By offering occupant seat assignment, dynamic UI modules, and remote operator escalation for rare conflicts, the system paves a path toward occupant-centric, globally scalable driverless fleets. We conclude that occupant concurrency logic, integrated with environment triggers and aggregator synergy, can transform fully autonomous ride-hailing from a purely technological achievement to a safe, user-friendly experience accessible worldwide.
Keywords:
Download/View Count: 105
Share this Article