Autonomous robots, equipped with the ability to perceive and understand their environment, hold immense potential to revolutionize industries ranging from manufacturing and logistics to healthcare and beyond. At the core of these autonomous systems lies the capability of object detection, enabling robots to identify and respond to various entities in their surroundings with precision and efficiency. The proposed system introduces an innovative approach capable of integrating YOLOv5 with the Robot Operating System (ROS) for the intended work and it is simulated using Gazebo simulation environment. The primary focus is on real-time object detection and obstacle avoidance, critical for enabling safe and efficient robot navigation in dynamic environments. The workflow entails initializing the robot within the Gazebo simulation environment, collecting and preprocessing sensor data, and applying the refined data to the YOLOv5 algorithm for obstacle classification and localization. A comprehensive object avoidance strategy is devised, incorporating advanced trajectory prediction algorithms such as A* and RRT. This integration offers promising prospects for advancing autonomy and adaptability in robotics, paving the way for the development of safer and more intelligent robotic systems.