ROS-Based Unmanned Mobile Robot Platform for Agriculture

While the demand for new high-tech technologies is rapidly increasing, difficulties are presented, such as aging and population decline in rural areas. In particular, autonomous mobile robots have been emerging in the agricultural field. Worldwide, huge investment is being made in the development of unmanned agricultural mobile robots; meanwhile with the development of robots, modern farms have high expectations of increased productivity. However, in the agricultural work environment, it is difficult to solve these problems with the existing mobile robot form, due to the difficulties of various environments. Typical problems are space constraints in the agricultural work environment, the high computational complexity of algorithms, and changes in the environment. To solve these problems, in this paper, we propose a method to design and operate a mobile robot platform that can be used in a greenhouse. We represent a robot type with two drive wheels along with four casters that can operate on path and rail. In addition, we propose a technology for a multi-AI deep learning system to operate a robot, an algorithm that can operate such a robot, and a VPN-based communication system for network and security. The proposed method is expected to increase productivity and reduce labor costs in the agricultural work environment.