Vehicle classification is a crucial topic in the image processing and computer vision research fields, where image classification is key field. The goal of the vehicle classification is to create a little system that can help with practical issues and uses, like traffic analysis, security, and autonomous and self-driving car technology. The conventional, labor-intensive methods of tackling image analysis difficulties have been surpassed which has greatly aided image processing-related issues. This paper used the symbiotic organism search approach to classify autonomous vehicles. An appropriately enriched dataset of car images from Kaggle was used to train the model. Several classifiers were used to classify the training model; however, one classifier was found to perform better than the others in terms of accuracy and time consumption. It has demonstrated to surpass similar model in terms of the accuracy and also in terms of the time and the speed predictions, according to the findings of empirical evaluations and statistical testing.