Fuzzy Logic Controller Design of a Single Stage Fluid Valve Based Robotic Arm

Abstract

Abstract One challenge for mobile robotics is the limitation of onboard processing power and the size of equipment required to accomplish basic tasks. While hydraulic pistons can generate high force output for various types of manipulation tasks, they require large equipment including the pistons, compressors, and the power sources in order to operate properly. For hydraulics to be applied to small to medium scale robots, the scale of the system needs to be as minimalistic as possible. For this purpose, we designed a single stage valve actuated robotic arm. Since system’s dynamic behavior is complex and only piston distance data is observable, a heuristic method, a Fuzzy Logic controller, is investigated. The goal is to create and optimize a system using a single piston and small compressor while maintaining effective control of the arm. Fuzzy Logic Controller system details, experimental setup and results are discussed for various motion profiles and finally its effectiveness is compared with a traditional PID controller approach. Robotics is a field that is constantly evolving and looking for creative new ways to provide solutions to new and old obstacles. One challenge for mobile robotics is the limitation of onboard processing power and the size of equipment required to accomplish basic tasks. The issue especially with smaller scale mobile robots is the lack of power and space on the robot to fit equipment and any critical apparatus for said equipment. While hydraulic pistons can generate high force output for various types of manipulation tasks, they require large equipment including pistons, compressors, and power sources in order to operate properly. Typically, a motor is used to power a pump which pressurizes hydraulic fluid. This fluid is used by the machine via hydraulic tubes and then is sent back through a filter and returns to the pump. This method of using hydraulics requires a large amount of power and a large amount of space. For hydraulics to be applied to small to medium scale robots, the scale of the system needs to be as minimalistic as possible. For this purpose, we designed a single stage valve actuated robotic arm. These employed the use of small electronic control boards fitted to the compressor system used to drive the hydraulics. For types of systems that must operate with discrete impulses of data to reach a target or reference value predominantly use Proportional-Integral-Derivative (PID) Controllers. These would be programmed in the onboard computer driving the hydraulic system. These systems require careful tweaking of various gain values to set up appropriately. However, the choice of an accurate controller structure relies on the application. Since system’s dynamic behavior is complex and only piston distance data is observable, an alternative method of control is investigated. Fuzzy logic is a control method which takes inputs from sensors and determines the output in terms of logical statements that can be in degrees between 0 and 1 rather than strictly 0 and 1 as seen in traditional Boolean logic. The particular equipment of interest in this study is a pneumatic robotic arm and all associated hardware. The goal is to create and optimize a system using a single piston and small compressor while maintaining effective control of the arm. Both a PID controller and a Fuzzy Logic controller were used to control the same system for the same set of tests. After PID control design approach is briefly discussed, Fuzzy Logic Controller system details, experimental setup and results are discussed for various motion profiles. Finally, the effectiveness of a traditional PID controller is compared to the effectiveness of a fuzzy logic controller.