The Research On Area-covering Control And Dynamic Characteristics For Intelligent Robot Mower

The intelligent robot mower (IRM) studied in this thesis is one kind of area-covering outdoor mobile robot working in the unstructured environments. With the full consideration of engineering practicability, based on the RBF neural network, a new solution to the boundary set-up and identification of unmarked operational area is proposed.The intelligence of robot mower is regarded as several adaptability behaviors in the environments, so the concept of unitary behavior mode is proposed and the unitary behavior-controlled architecture for robot mower is formed. The characteristics of the unitary behavior mode are deeply studied. We put forwards the behavior-unit SB and SDB models, and use the Fuzzy Petri Net to build the environment-oriented and task-aimed behavior model for IRM. On the internal relationships between the behavior control and fuzzy logic, we design the main behaviors for IRM covering the whole operational area by the use of the fuzzy ratiocination theory. That is: path-tracing behavior, obstacle-avoiding behavior and region-switching behavior. The harmonization problem of path-tracing behavior and obstacle-avoiding behavior is solved by the fuzzy control.According to the purposes and features of IRM, the evaluation indexes of robot performance are given. The mathematics description of complete coverage path planning is presented. The dividing method of operational sub-regions is given based on the Morse theory, and the uniform and simple coverage paths can be planned. Combined with the boundary of operational area and the topology structure, a new method of embroiled sub-region with priority-covering is presented to obtain the motion planning more stably and continuously. The performance analysis of the complete coverage planning is given in the thesis.It is the first time that the research of aerodynamics for IRM is introduced. We analyze detailedly the flow mode and characteristics of airflow in the blade deck, and study the effect on IRM's self-acting behaviors produced by the imbalance airflow. By the solution of the finite element model, we also do the analysis for the dynamic response of centrifugal inertia force from the imbalance blades impaction on the body of IRM, and eventually put forward the method to decrease the forced vibration.The dynamic model of IRM in the lawn environment is build up, the related analysis of dynamic characteristics based on the model is studied. Based on the physical feature of lawn, by investigating the particular mechanics property of wheel-lawn model, the rolling resistance and the allowable attaching force are obtained when the robot mower moves on the lawn. For the dimensional feature of lawn, the function is set to describe the lawn's unevenness. The vertical vibration of IRM is studied based on the lawn terrain function, and then the vibration characteristics will be available at the different velocities of the robot motion. Through studying the steady response of lateral disturbance from the uneven lawn, the steady response will be obtained.Using the Backstepping method, we design the trajectory following law based on the Nonholonomic dynamics of robot mower. The emulation results considering the disturbance are given. The feedback Stabilization control of robot is discussed using the H_∞ theory.Considering the demand of the environment and tasks, we design the mechanical structure and control system based on the CAN field bus technology. The main module...