Hardware Development Of Drag-and-drop Multi-joint Teaching Robot Learning Platfom Of Position And Orientation

With the continuous expansion of the application field of industrial robots, it is very important for robot to realize fast programming, reduce auxiliary time and improve production efficiency. At present, the commonly used programming approach is manual teaching. Although this method can accurately find trajectory, it is hard to realize the complex and changeable movement state. And off-line programming teaching which requires high professional knowledge for technical staff is difficult to achieve. Drag-anddrop teaching with merits of celerity, intuition and high efficiency has a very broad application prospects in the field of industrial robot spraying. A set of drag-and-drop multi-joint teaching robot learning platform of position and orientation is developed in this thesis and the robot's joint information is collected to generate the robot's trajectory. The main works of this thesis are as follows:Based on full study of the current developing situation of teaching technology at home and abroad, strengths and weaknesses of various teaching techniques are summarized, and the teaching scheme of drag-and-drop robot learning platform is proposed. Combined with the characteristics of robot drag teaching, the structure of the whole learning platform and the overall plan of hardware are designed.The hardware platform of the system is designed in detail. Altera's Cyclone II FPGA is chosen as the main processor of the signal acquisition system. Schemetic diagrams of power supply system, encoder feedback as well as RS485 interface are completed. While developing storage system, with TI's AM335 x as the core CPU, minimum system of which is accomplished and the peripheral USB interface program is analyzed. To guarantee the stability and reliablity of hardware system, signal integrity analysis and simulation of the encoder interface circuit are studied and the specification of PCB stack and layout is reasonably regulated.Applying modular design concept, FPGA programs which involves drag-and-drop robot pose learning Tamagawa encoder protocol, RS485 protocol and communication between ARM and FPGA are designed through using VHDL. Otherwise, the data storage solution of USB flash disk in embedded system is developed.The test platform is set up and each module's function is tested separately. Signal of each joint is collected by dragging the teaching robot. The angle information acquired then is transformed into the pose position and orientation information. Finally, the trajectory fitting is achieved in MATLAB and the result is satisfactory.