Development of Service Robot Manipulator for Pick and Place applications in Domestic Environment

Service robots have been developed to assist humans in daily tasks, particularly in environments where repetitive or complex manipulations are required. Robotic manipulators have previously been studied mostly for factory or organized applications, usually with a lack of flexibility, intuitive control, or adaptability for general-purpose service tasks. This project specifically focuses on applications within domestic environments, addressing those limitations by developing a service robot manipulator capable of both autonomous object handling and manual control via a user-friendly interface. The study was conducted in two phases: a simulation environment and a physical prototype. In the simulation phase, object detection was performed using the YOLOv8 Object Box Bounding model. The detected object positions were calculated. Further robotic path planning and execution were carried out using ROS and MoveIt frameworks. In the prototype phase, an ESP32- based manipulator was built using servo motors. The ESP32 hosted a Wi- Fi access point and served a web-based interface where users could manipulate the arm through sliders and control buttons. Real-time bidirectional communications was established via WebSocket, allowing the user to record motion sequences and replay them with accurate timing. Results from both simulation and prototype implementation demonstrate successful grasping, motion planning, and repeatability of tasks. This study contributes to existing research by presenting a modular, low-cost, and interactive system that can be applied in various service applications, from home assistance to small-scale automation.