Mobile Robot In The Greenhouse Environment Map Of The Environment To Build

Mobile robot is an important branch of robotics, and navigation technology is the core. Autonomous mobile robots have been widely used in industry. And it also attracts scholars in the field of agricultural engineering. Agricultural robots is a top research in agricultural engineering, There have been picking robot and the autonomous navigation of agricultural vehicles.As for indoor small-scale environment, we can not use of existing technology to navigatation.As a resulat, we propose a map-based navigation methods, solving the agriculture mobile robot navigation problem in small-scale structure or semi-structured environment, The main work of this paper is given as follow:1 Based on extensive reading we get a conclution that Mobile robot navigation problem can decompose into three basic problems, namely: map building, robot localization, and path planning. The paper discusses these three problem and the cross-cutting issues were discussed.2 Aim at the structure of the greenhouse environment, we using the grid method, to construction grid map. With ultrasonic sensors as the main sensing element, the establishment of ultrasonic sensor model. And analysis source of error. We use Classic D-S evidence theory to fusion sonar information. The paper proposes an improved DS evidence theory.And use MATLAB to make the simulation results.3 Using threshold-based segmentation to constructt the global environment, the method need great different between the background and the obstacles the environment. Mapping accuracy depends on the size of the grid partition, and use MATLAB to simulation results.4 For the robot localization problem. We use HSI color space to determine the location information of mobile robot, and make use of MATLAB simulation results.5 To solove the path planning problem. We were using the ant colony algorithm to plan paths. The algorithm can not find the global optimal path for mobile robot. However, the algorithm is simple, rapid convergence. And can quickly search a sub-optimal path in practical work.