A method and system of adaptive spatiotemporal synchronization for multi-source data in intelligent vehicle-infrastructure collaborative perception

Abstract

Provided in the present invention are a multi-source data spatial-temporal adaptive registration method and system oriented to road-vehicle intelligent fusion perception. The registration method comprises: a connected vehicle carrying a vehicle-mounted positioning sensor and a vehicle-mounted perception sensor driving in the sensing range of each road-region sensor to collect calibration data; on the basis of the sensed data of the vehicle, by means of discrete trajectory data continuity, weighted feature map creation and map matching similarity matrix calculation, determining a vehicle-road trajectory data matching relationship; and, on the basis of the position data of the vehicle, successively solving calibration parameters of a single sensor and a plurality of sensors to implement multi-source data adaptive registration. On this basis, the multi-source data spatial-temporal adaptive registration system oriented to road-vehicle intelligent fusion perception provided by the present invention comprises three modules, i.e. a vehicle end apparatus module, a road-region apparatus module and a multi-source data spatial-temporal adaptive registration platform. Compared with the prior art, the present invention may implement rapid and automatic spatial-temporal adaptive registration of the road-vehicle intelligent fusion perception multi-source data, and implement efficient spatial-temporal synchronization of the road-vehicle intelligent fusion perception multi-source data.

Dependencies are inferred from the claim text.