The extraction of vibrations allows failure detection [20] and some intelligent sensors have been proposed to perform this task [14]. Accelerometers have been also included in servo control loops [12]. By analyzing vibrations in combination with force sensors, contact forces are measured and calibrated [13], manifesting the relevance of a proper separation of vibration signals from the raw accelerometer measurement. Moreover, static acceleration indicates the inclination of the accelerometer with respect to gravity and by taking these signals, kinematics calibration in a manipulator arm can also be performed [11]. Inclination parameters have been investigated utilizing an accelerometer as primary sensor [21].
Furthermore, an integrated approach utilizing an encoder and an accelerometer has been presented to accurately estimate velocity [8].
Additionally, sensors are becoming more intelligent by integrating signal conditioning, processing units, communication protocols, among other features [3]. Therefore, the development of a smart sensor that integrates data fusion of motion dynamics, inclination, and vibration parameters is considered an essential move towards intelligent-robotics.This Entinostat work presents a novel smart sensor that extracts motion dynamics and inclination parameters along with the separation of vibration information from a single link in industrial robots, based on the fusion of two primary sensors: an optical incremental encoder and a triaxial accelerometer.
Motion dynamics is estimated from the encoder measurement to give position, velocity, acceleration, and jerk; whereas vibrations and inclination are separated from the accelerometer signal, for providing angular position, velocity, acceleration, and vibrations. Estimated parameters are computed online utilizing digital Carfilzomib hardware signal processing techniques such as digital filtering, interpolation, finite differences, among others. These computer-intensive processing algorithms are implemented in a field programmable gate array (FPGA) for a smart sensor approach by integrating hardware signal processing and data communication in an embedded system.2.
?BackgroundThis section establishes the relationship among the estimated parameters on a robotics application, where the encoder gives information regarding motion dynamics and the accelerometer gives both inclination and vibration information.2.1. Motion DynamicsMotion in a manipulator arm is conducted by the motion controller that applies a profile to perform smooth movements to the end effector. By taking the position feedback signal (p) from the servo control loop, velocity (v), acceleration (a), and jerk (j) can be estimated.