Developed a vehicle speed prediction model for long and short duration using deep learning, machine learn-ing, probabilistic and stochastic models by acquiring real time traffic and map data from different companies.
Used LSTM model for time series data prediction and deep probabilistic programming using Bayesian NeuralNetworks and approximate Bayesian LSTM for determining its confidence interval.
Worked with different R&Ds and MDOT to develop a DSRC module for vehicle to vehicle and vehicle to infrastructure communication and to transmit messages over CAN bus.
Performed real time validation and testing on road to evaluate the performance of the speed prediction model and DSRC data communication with traffic lights.
Published a SAE technical paper along with other researchers on various artificial intelligence approaches to predict vehicle speed for realizing Predictive Powertrain Control.