Vehicular Ad-hoc Networks (VANETs) has changed the transportation environment in a big way through car-to-car (V2V) and car-to-road Vehicle-to-Infrastructure (V2I) communications. In VANETs, good route planning plays an indispensable role in achieving a safer, more efficient transportation system. This project proposes a novel deep learning-based approach for next-generation route planning algorithms in VANETs. Using the power of deep learning, our method tries to achieve optimal route planning in VANETs, decreasing travel time, and emissions whilst increasing total throughput of transportation. Our proposed algorithm combines on-road traffic information and the road conditions in the outdoor environment with some environmental factors to forecast optimized routes in the dynamic VANET environments. The presented method is tested in an extensive simulation environment, which illustrates the potential of the proposed method to substantially enhance the effectiveness and the safety of route planning in VANETs. These studies have important potential applications for the advancement of intelligent transportation systems, smarter and greener urban mobility.