A contemporary autonomous car selects its driving strategy based solely on outside influences (In terms of foot traffic, state of roads, and so on), ignoring internal state of vehicle. This work offers “Driving Decision Strategy with using a stacking approach for an independent vehicle" which decides ideal technique of an independent vehicle by breaking down not only exterior aspects, though additionally internal state of vehicle to solve problem (consumable conditions, RPM levels, etc.). DDS builds an evolutionary algorithm and produces the best driving strategy for an automated car based on data from automotive sensors saved in the cloud. In order to verify DDS, this study compares it to two other neural network models, the multi-layer perceptron (MLP) and the recurrent fuzzy network (RF). When compared to conventional vehicle gates, DDS's loss rate was about 5% lower, and it was 40% quicker at determining RPM, speed, turning angle, and position shifts.