In this paper, we present an incentive mechanism for vehicular crowdsensing (VCS) based on a recurrent reverse auction. The proposed approach encourages participant's vehicles to used their sensors to collect data while also maximizing their utility. This approach tackles important issues in VCS such as cost explosion, and area sensing coverage. Using a realistic street network from OpenStreetMaps with extensive SUMO (Simulation of Urban Mobility) simulations, we show our VCS algorithm significantly o utperforms the baseline approach in terms of sensing coverage and active number of participants by three and eight times respectively.