A Brain-Computer Interface (BCI) allows a person to manipulate a system entirely by thoughts that originate in the brain rather than using any physical limbs. These interfaces are designed to assist in many applications, especially rehabilitating paralyzed people. In this paper, we present a BCI interface to classify and control seven movements of a wheelchair; forward, backward, left, right, stair climbing upwards, stair climbing downwards, and stop. The electroencephalography (EEG) technique collects raw signal data from neurologically healthy volunteers. We initially pass the captured signals through filtering before feeding them to the feature extraction and classification stages. We test three classification algorithms to evaluate our approach: Convolution Neural Network (CNN), Support Vector Machines (SVM), and Random Forest Classifier. We present a comparative analysis of these classifiers for better results. Experimental results showed that the proposed approach was promising for implementing BCI with a classification accuracy of 99\% using a Random Forest Classifier.