From Traditional to Autonomous Vehicles: A Systematic Review of Data Availability

The increasing accessibility of mobility datasets has enabled research in green mobility, road safety, vehicular automation, and transportation planning optimization. Many stakeholders have leveraged to study conventional driving characteristics self-driving tasks. Notably, many these been made publicly available, fostering collaboration, scientific comparability, replication. As encompass several domains contain distinctive characteristics, selecting the appropriate dataset investigate aspects might be challenging. To best authors’ knowledge, this is first paper that performs a systematic review substantial number covering various automation levels. In total, 103 reviewed, 35 which focused on naturalistic driving, 68 gives researchers possibility analyzing datasets’ principal their domains. Most centered safety driver behavior, although eco-driving also studied. Furthermore, for autonomous analyzed according target A particular focus placed data-driven risk assessment ecosystem. It observed there exists lack relevant available challenge creation new models semi- fully automated vehicles. Therefore, conducts gap analysis identify possible approaches using existing and, additionally, set data fields could incorporated future collection campaigns address challenge.