Research on simulated visual control of robotic arm in ROS environment

Traditional robotic arms often lack sufficient perceptual capabilities and flexibility in dynamic working environments. To address this issue, this paper proposes integrating visual sensors into the robotic arm system to ensure real-time scene information acquisition, thereby enhancing the real-time operational capabilities of the robotic arm. Built a simulation model for visually guided robotic arm in the ROS environment. Conducted a joint simulation using Rviz and Gazebo, completing the modeling and simulation of the robotic arm and MV-DB500S-R camera. Implemented the guidance of the robotic arm's grasping path based on visual recognition results through the ROS node communication mechanism. The effectiveness of the proposed method was verified through simulation results.