In this paper, we introduce a mobile robotic system with a 7-Degrees of Freedom Robotic Arm designed for executing various pick-and-place operations and autonomous navigation within indoor environments. This system integrates vision-based object recognition and localization to facilitate precise grasping actions. A visual simultaneous localization and mapping (SLAM) algorithm is utilized to enable the robot’s autonomous navigation in unfamiliar indoor settings. Additionally, we employed the YOLOv3 algorithm which enhances object detection and tracking capabilities, which is crucial for responding to any sudden positional changes in the target object. The process begins with YOLOv3 identifying and locating the object, followed by monitoring the next frame to detect any immediate shifts before the robotic arm’s movement to grasp the object. Once the object to be picked up is decided by the Robot, we employ a Countour-based Object boundary detection technique that gives us both, the objects’ coordinates in 3D space and the orientation at which the object is placed. Subsequently, a coordinate transformation is performed, converting the coordinates of the target object from the camera’s frame of reference to that of the robotic arm’s base frame, thereby facilitating accurate grasping operations. Furthermore, the robot is equipped with a human-following algorithm, enabling it to track and transport payloads to a human target efficiently, placing the payload in proximity to the person. Experimental results highlight a significant grasping success rate of 94%, underscoring the system’s efficacy in autonomous robotic navigation and pick-and-place operations. This advancement holds substantial promise for enhancing robotic autonomy and operational efficiency in real-world scenarios.