Indoor spraying is a crucial component of the construction industry. The future evolution of the spraying industry will inexorably involve the substitution of robot-assisted autonomous spraying for manual spraying. In this paper, a fully autonomous indoor spraying robot is designed to handle a variety of manual spraying issues. At the system level, the robot’s mechanical structure, hardware system, general control logic, and crucial technologies are meticulously constructed. The robot is equipped with a chassis that can travel in any direction, a detachable material carrier, a secondary upgrade mechanism, and a spraying system. The robot performs indoor mapping and navigation with the lidar sensor, identifies non-sprayable areas, such as windows, using the camera, and then evaluates the spraying result after spraying. After analysis of the robot’s working environment, the SLAM algorithm and the deep learning-based object detection algorithm are improved in conjunction with the actual scene to assure accuracy and meet the criteria for real-time operation on embedded devices. The robot’s ability to perform a succession of indoor spraying activities without human intervention and to automatically adapt and adjust to varied indoor conditions is an important reference for its practical application in the field of interior design.

室内喷涂是建筑行业的重要组成部分。喷涂行业未来的发展将不可避免地涉及机器人辅助自主喷涂取代手动喷涂。本文设计了一种完全自主的室内喷涂机器人来处理各种手动喷涂问题。在系统层面，精心构建了机器人的机械结构、硬件系统、通用控制逻辑和关键技术。该机器人配备可任意方向行驶的底盘、可拆卸的物料载体、二次升级机构和喷涂系统。机器人通过激光雷达传感器进行室内测绘和导航，通过摄像头识别窗户等不可喷涂区域，然后在喷涂后评估喷涂效果。通过对机器人工作环境的分析，结合实际场景改进SLAM算法和基于深度学习的物体检测算法，保证精度并满足嵌入式设备实时运行的标准。该机器人能够在无需人工干预的情况下执行连续的室内喷涂活动，并自动适应和调整不同的室内条件，这对其在室内设计领域的实际应用具有重要参考意义。