arXiv:2605.29768v1 Announce Type: new Abstract: Existing traffic forecasting benchmarks assume a fixed sensor set, but real road-sensor networks grow continuously as the road network changes year by year. We introduce the XXLTraffic dataset family, which spans up to 27 years of California PeMS and Transport for NSW data. The fixed-sensor subsets of XXLTraffic support extremely long forecasting with multi-year gaps and standard hourly / daily long-horizon forecasting. We extend it to EvoXXLTraffic, a sensor-evolving reorganization that exposes per-year active sensors, yearly traffic-flow matric
The proliferation of IoT sensors and the increasing complexity of urban environments necessitate more robust and adaptive AI models for real-time traffic management.
This development addresses a critical gap in existing AI for real-world infrastructure management by accounting for dynamic sensor networks, improving predictive accuracy and operational efficiency.
Traffic forecasting models can now adapt to evolving sensor infrastructures rather than relying on static datasets, paving the way for more resilient and accurate smart city applications.
- · Smart City technology providers
- · Urban planners
- · Logistics companies
- · AI model developers
- · Legacy traffic management systems
- · Companies relying on static data approaches
Improved urban traffic flow reduces congestion and commute times.
Better traffic predictions lead to optimized route planning and decreased fuel consumption and emissions.
The methodology could be extended to other dynamic sensor networks, such as environmental monitoring or utility grids, enhancing AI resilience across critical infrastructure.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI