رابط انسان و منبلترس جدید با استفاده از سنسور ترکیبی با فیلتر کالمن و فیلتر ذرات

This paper presents a novel method for a human–manipulator interface, which estimates the position and orientation of the human hand using a 3D camera and an inertial measurement unit (IMU). In the proposed method, a 3D camera is used to locate the human hand with the help of Camshift algorithm and an IMU is employed to measure the orientation of the human hand. Although the position and the orientation of the human hand can be obtained from the two sensors, the measured error increases over time due to the noise of the devices and the tracking error. Particle filter (PF) and Kalman filter (KF) are used to estimate the orientation and the position of the human hand. Finally, a finite impulse response (FIR) filter is used to detect the PF failures. The human hand is mapped to the robot EE so that human could control the robot just like controlling his/her own hands. The greatest advantage of this method is that the human can complete the high precision task without much training beforehand. The human–robot interface system was experimentally tested in a lab environment and the results indicate that such a system can successfully control robot manipulator.