Carlos Rodriguez-Donate

FPGA-Based Fused Smart Sensor for Dynamic and Vibration Parameter Extraction in Industrial Robot Links

Intelligent robotics demands the integration of smart sensors that allow the controller to efficiently measure physical quantities. Industrial manipulator robots require a constant monitoring of several parameters such as motion dynamics, inclination, and vibration. This work presents a novel smart sensor to estimate motion dynamics, inclination, and vibration parameters on industrial manipulator robot links based on two primary sensors: an encoder and a triaxial accelerometer. The proposed smart sensor implements a new methodology based on an oversampling technique, averaging decimation filters, FIR filters, finite differences and linear interpolation to estimate the interest parameters, which are computed online utilizing digital hardware signal processing based on field programmable gate arrays (FPGA).

FPGA-based fused smart-sensor for tool-wear area quantitative estimation in CNC machine inserts.

Manufacturing processes are of great relevance nowadays, when there is a constant claim for better productivity with high quality at low cost. The contribution of this work is the development of a fused smart-sensor, based on FPGA to improve the online quantitative estimation of flank-wear area in CNC machine inserts from the information provided by two primary sensors: the monitoring current output of a servoamplifier, and a 3-axis accelerometer. Results from experimentation show that the fusion of both parameters makes it possible to obtain three times better accuracy when compared with the accuracy obtained from current and vibration signals, individually used.

FPGA based embedded system for induction motor failure monitoring at the start-up transient vibrations with wavelets

Accurate monitoring on induction motors is mandatory for modern industry in order to guarantee the overall process quality. The common practice for monitoring is performed by third party enterprises that test the electrical machines with general-purpose instrumentation equipment, which do not allow on-line operation with the subsequent increase in production costs. Several methodologies have been proposed in recent years for detection of failures in induction motors; however, these methodologies perform the analysis offline. The contribution of this work is the development of an online monitoring of induction motor failures by measuring the vibration transient signals at the start-up with discrete wavelet transform, and its implementation as an embedded system with FPGA for SOC approach. Experimentation is realized to test the system functionality. From results it is demonstrated that the proposed methodology accurately determinates the motor condition in the presence of broken rotor bars.

Wavelet-based general methodology for multiple fault detection on induction motors at the startup vibration transient

Fault preventive monitoring on induction motors has risen in order to reduce maintenance costs and increase their life expectancy. There are many developments for detecting a single induction motor fault using several methodologies and techniques. Different methodologies have been developed for multiple fault detection having the disadvantage of giving a qualitative result requiring an expert technician for estimating the motor condition, with the possibility of inducing observation errors. This work proposes a quantitative general methodology for online induction motor monitoring and identification of multiple faults in an automatic way, and its hardware processing unit for real time applications, based on the startup vibration transient analysis. The proposed methodology is tested on three different cases of study: a motor with broken rotor bars, an unbalanced motor shaft, and a motor with misaligned load. The results show that the proposed methodology is highly reliable for detecting different faults in induction motors with a certainty of 99.7%. The developed approach can be extended for detecting other faults by a proper calibration, thanks to its generalized nature.

Fused Smart Sensor Network for Multi-Axis Forward Kinematics Estimation in Industrial Robots

Flexible manipulator robots have a wide industrial application. Robot performance requires sensing its position and orientation adequately, known as forward kinematics. Commercially available, motion controllers use high-resolution optical encoders to sense the position of each joint which cannot detect some mechanical deformations that decrease the accuracy of the robot position and orientation. To overcome those problems, several sensor fusion methods have been proposed but at expenses of high-computational load, which avoids the online measurement of the joint’s angular position and the online forward kinematics estimation. The contribution of this work is to propose a fused smart sensor network to estimate the forward kinematics of an industrial robot. The developed smart processor uses Kalman filters to filter and to fuse the information of the sensor network. Two primary sensors are used: an optical encoder, and a 3-axis accelerometer. In order to obtain the position and orientation of each joint online a field-programmable gate array (FPGA) is used in the hardware implementation taking advantage of the parallel computation capabilities and reconfigurability of this device. With the aim of evaluating the smart sensor network performance, three real-operation-oriented paths are executed and monitored in a 6-degree of freedom robot.

Early Experiences with OpenCL on FPGAs: Convolution Case Study

Many proprietary standards and tools have been designed in order to cover a closed set of architectures, and OpenCL has become a free standard for parallel programming on heterogeneous systems, which include custom devices, CPUs, GPUs, FPGAs. This work evaluates the use of the well-known convolution operator in signal processing disciplines focused on FPGA evaluation under different optimizations with respect to thread and memory level exploitation.

FPGA based multiple-channel vibration analyzer embedded system for industrial applications in automatic failure detection

Machine monitoring is one of the major concerns in modern industry in order to guarantee the overall efficiency during the production process. Several monitoring techniques for machinery failure detection have been developed, being vibration analysis one of the most important techniques. The typical equipment used for vibration analysis is a general purpose single channel spectrum analyzer that most of the cases is not well suited for the specific task and lacks from the capability of simultaneous multiple channel analysis. The contribution of this work is to present the development of a low-cost FPGA based 3-axis simultaneous vibration analyzer for embedded machinery monitoring with the novelty of a post-processing stage that can be designed and implemented into the same FPGA for automatic online detection of specific machinery failures. Two cases of study are presented to show the development performance and capabilities of the system where specific post-processing units are designed. From the results it can be seen that several mechanical failures can be automatically detected by reconfiguring the postprocessing algorithm, embedded in the system.

Novel hardware processing unit for dynamic on-line entropy estimation of discrete time information

Abstract This paper presents a novel and generic hardware processing unit that estimates the information entropy in a dynamic and on-line fashion with a simple architecture that can be easily scaled. This architecture does not require precomputations, change of domain at the input signal, or complex schemes of computation. Results show that the proposed FPGA implementation of the dynamic entropy estimator is highly efficient as a stand-alone system. Speed performance of the system is 3 orders of magnitude higher than its implementation counterpart in software with a maximum error of 1.5%. Compared with other hardware structures, the proposed architecture is able to process twice the information than a LUT-based entropy estimator during a time unit. Results also show that the proposed dynamic hardware processing unit is highly accurate carrying out standard tasks such as computing the information content in a discrete data set, or nonstandard tasks as detecting failures in induction motors.

Novel FPGA-based Methodology for Early Broken Rotor Bar Detection and Classification Through Homogeneity Estimation

Early detection of induction-motor faults has been an increasing matter of research in the last few years. The reliable identification of broken rotor bars (BRB) is still under investigation as it is one of the most common and difficult-to-detect faults in induction motors. Many methods have been proposed to deal with this issue. Recent approaches combine techniques looking for improving the performance of the diagnosis. Their major disadvantage is the high computational requirements, which restrains them from being used in online detection. The contribution in this paper is twofold. The first one is a novel methodology for induction motor BRB detection and the fault severity classification using homogeneity as index, which, to the best of our knowledge, has never been used as an indicator for fault diagnosis, analyzing one phase of the induction motor startup-transient current. Because of the low computational complexity in homogeneity calculation, the second contribution of this paper is a hardware-processing unit based on a field programmable gate array device for online detection and classification of BRB. Obtained results demonstrate the high efficiency of the proposed methodology as a deterministic technique for incipient BRB diagnosis in induction motors, which can detect and differentiate among half, one, or two BRBs with a certainty greater than 99.7%.

FPGA-based reconfigurable unit for image encryption using orthogonal functions

Encryption is an important tool in many areas of application and research. The advances in communications have encouraged researchers to find new techniques for providing data security, confidentiality, integrity and authentication. The techniques proposed until now for image encryption apply well-known image processing techniques, increasing their computational complexity and processing time, threatening their use on real-time applications. On the other hand, the already proposed hardware implementations for image encryption do not allow portability to distinct FPGA platforms, and they do not guarantee high speed and optimal resource utilization. In this work, a generic real-time, FPGA-based, reconfigurable architecture for online image encryption using orthogonal functions is proposed. The introduced architecture implements a novel highly-efficient algorithm for high speed image encryption using minimal resources, which is portable to different FPGA platforms from distinct vendors. Obtained results demonstrate the effectiveness of the proposed approach on different cases of study, reaching processing rates up to 39 frames per second.