Robotic Systems and Applications
Not a member yet
20223 research outputs found
Sort by
Making humanoid robots teaching assistants by using natural language processing (NLP) cloud-based services
Get PDFHumanoid robots have a substantial potential to serve as teaching and social assistants. However, the expectations of the children from robots to interact like humans are huge. This study presents a general model for understanding the natural language in human-robot interaction by applying Generative Pre-trained Transformer (GPT) language models as a service in the Internet of Things. Thus, the physical presence of the robot can help in fine-tuning the GPT model by prompts derived from the environmental context and subsequent robot actions for embodiment understanding of the GPT outputs. The model uses web or cloud services for Natural Language Processing (NLP) to produce and play human-like text, question answering or text generation. Verbal questions are processed either via a local speech recognition software or via a Speech-to-Text (STT) cloud service. The converted question into machine-readable code is sent to one of the GPT language models with zero- or few-shot learning prompts. GPT-J model has been tested and deployed either in the web or cloud with options for varying the parameters for controlling the haphazardness of the generated text. The robot produces human-like text by using Text-to-Speech (TTS) cloud services that convert the response into audio format rendered on the robot to be played. Useful requirements how the model to be used in order to be feasible were determined based on the evaluation of the outputs given from the different NLP and GPT-J web or cloud-services. We designed and implemented the model in order to be used by a humanoid NAO-type robot in the speech language therapy practice, however it can be used for other open-source and programmable robots and in different contexts
Traveling wave effect of multi-span through concrete-filled steel tubular arch bridge
Get PDFTo explore the different influence of traveling wave effect on the isolated and non-isolated arch bridge, the isolated and non-isolated multi-span arch bridges models are established respectively, three measured seismic waves were selected, and under the eight kinds of apparent wave velocities and multi-point consistent excitations, the structural responses of the two models, including the internal force response, the arch rib velocity, the pier’s internal force response, the bridge deck acceleration, and the shear force and displacement of isolation support of different arch ribs position in two models, were compared and analyzed. The results show that the isolation effect of the isolation structure is obvious; the wave effect of the isolated structure is significant, the two structures show different internal force response curves in different positions of the arch ribs under the influence of different seismic waves; the influence of the apparent wave velocity on the force and shock reduction rate of the arch ribs is complex; the vertical acceleration of the arch ribs and bridge deck of the isolated structure is reduced significantly; and the shear force and displacement of the isolation bearing increase with the increase of the apparent wave velocity. The study shows the traveling wave effect of multi-span through a concrete-filled steel tubular arch bridge with and without isolation, and the results will be used for the seismic design and analysis of structural diseases caused by the wave effect
Detection of weak joints and damages for beams using machine learning
Get PDFFor maintaining the safe operation of structures, it is necessary to develop SHM methods that can detect not only the presence of cracks in the structure but also any alterations of its fastening conditions. The current paper presents a method for developing an Artificial Intelligent model that can detect if a beam is affected by transverse cracks and at the same time, by improper boundary conditions. To this aim, a cantilever steel beam is considered as the in the current study. The training data for the artificial neural network (ANN) is created using an original analytic method which allows calculating the natural frequency loss caused by the occurrence of transverse cracks even if the beam is improperly fastened. The intelligent model is trained by employing the MATLAB software and tested using data acquired from numerical simulations. The results show very high accuracy in determining the presence of transverse cracks, and the capability of detecting the presence and severity of improper clamping conditions
Determining the position of two cracks in a cantilever beam using artificial neural networks
Get PDFDuring functioning time, structures such as beams are subjected to a variety of loads caused by the working conditions and environment, which can lead to the development of cracks. The current research is concerned with detecting the presence and location of two transverse breathing cracks, in cantilever beams. Starting from the known fact that damages produce a stiffness degradation in structures altering their dynamic parameters, we performed modal simulations of damaged beams to determine their natural frequencies. By using the simulation data, we trained an artificial neural network (ANN), using the feedforward backpropagation algorithm, that is capable to detect the presence of the cracks, their position and for the case when the damages are in proximity, the model can determine if the cracks occur on the same face or opposite faces
Simulation analysis on seismic dynamic response of pile supported tunnels in deep backfill area of soil-rock mixture
Get PDFTo reveal the seismic dynamic response of the pile-supported tunnel group in the soil-rock mixture deep backfill region, a three-dimensional finite element model was established based on the engineering conditions of the subway section and three tunnels with close access lines. Subsequently, the seismic dynamic response of the tunnel lining structure was studied. The results show that: Under the action of seismic, the soil-rock mixture stratum presents nonlinear characteristics with shear failure and plastic deformation. In addition, the acceleration and earth pressure of the soil-rock mixture stratum is in a “saturated” state; The seismic dynamic response of the three tunnels influences each other. The bending moments in the X and Y directions of the tunnel lining cross-section are distributed in “X” and inverted “V” shapes, respectively. Meanwhile, the tensile stress and shear stress are distributed in an “X” shape; Under the action of seismic, the main failure form of tunnel lining is tension shear failure, and the most vulnerable position is the left and right arch foot, followed by the left and right arch shoulder; The bending moment of the pile body changes nonlinearly in the height direction. The most significant bending moment value appears at the top 1/5 of the pile length and the junction of different strata. Furthermore, the most significant horizontal displacement of the lining structure occurs at the tunnel vault
Prediction of operation vibration state of coal mine mechatronics equipment based on data mining
Get PDFIn order to explore the prediction of the operation vibration state of coal mine mechatronics equipment, the author proposes a method based on data mining in response to the problems of large amount of data in the operating state of coal mine electromechanical equipment, low data utilization, and slow speed of single-machine massive data mining, using Map Reduce technology, a dual Map Reduce mining prediction framework is proposed, establish a data mining prediction model for the running state of dual Map Reduce, using MapReduce1 to extract features of monitoring data, use MapReduce2 to predict and analyze feature data. Finally, by building the Hadoop platform, reveal the relationship between Hadoop cluster nodes and parallel processing speed, the efficiency of the data mining prediction framework is verified: Perform an experimental comparative analysis of the single prediction model and the proposed AGB combined prediction model, the prediction accuracy of the AGB combined prediction model is verified
Diagnosis and analysis of electric power steering rattle noise based on multiple coherence method
Get PDFElectric power steering (EPS) system is becoming the primary source of Electric Vehicle (EV) noise due to absence of classic power train. The rattle noise produced by EPS, which is easily discernible when driving. Since the conventional subjective evaluation approach makes it difficult to identify the source, it is required to establish an objective way to solve such an issue. In this paper, the principle of multiple coherence method is first addressed analytically, and then is employed to analyze the vibration signal acquisition from EPS different points, meanwhile, combine with in-car sound signal, to accurately identify the specific location
Fault identification and remaining useful life prediction of bearings using Poincare maps, fast Fourier transform and convolutional neural networks
Get PDFBearings are integral components of rotating machinery and their failure tends to be a catastrophic failure of the machine. Poincare Maps are used to detect bearing failures using the concept of non-linear dynamics. Each time-domain vibration signature array has its own Poincare Map over a period of time. Fast Fourier Transform (FFT) is a method of analysing the frequency plots of a bearing signature. Convolutional Neural Networks (CNN) process the bearing Continuous Wavelet Transform images and provide the Remaining Useful Life (RUL) of the bearing. The Poincare Maps and FFT plots are used to diagnose the type and location of the fault in the bearing, whereas the CNN helps to provide the fraction of Remaining Useful Life. The study concludes that a combination of Poincare Maps, FFT analysis and Convolutional Neural Networks constitutes a robust and precise method of monitoring bearing conditions
Effect of defect distribution on thermal expansion coefficient of eutectic composite ceramics
Get PDFBased on Eshelby’s equivalent inclusion theory, the four-phase model and the interaction direct derivative estimate, the prediction model of effective thermal expansion coefficient of composite containing multiple types of inclusion in anisotropic matrix was established. The effective thermal expansion coefficient of eutectic composite containing defects was calculated. And then the influence of defects and inclusions on the effective thermal expansion coefficient is discussed in detail. The results show that the influence of inclusions will be amplified by defects when there are multiple inclusions in the matrix. Therefore, the interaction direct derivative estimate cannot accurately predict the influence of defect distribution on thermal expansion coefficient of eutectic composite ceramics
Forecasting rainfall and potential for repeated events to predict flood areas in Banten province, Indonesia
Get PDFIn a period of ten years, from 2011-2020 rainfall in Indonesia is relatively high, with annual rainfall between 460.5-4,627.4 mm. The high rainfall has implications for flooding in several provinces. During this period, almost every year several areas in Banten Province experienced floods. To predict areas of Banten Province that have the potential for flooding, forecasts of rainfall and the potential for repeated occurrences of high rainfall are carried out. In making the forecast, observations were made at the Serang Meteorological Station, the Budiarto Curug Meteorological Station, the South Tangerang Climatology Station, and the Tangerang Geophysics Station. Rainfall data from the four stations were analyzed by Fourier transform, Gumbel method and Mononobe method. Distribution analysis results obtained rainfall in Banten Province between 0.0-607.9 mm with the length of rainy days per month between 0-26 days. Then, the results of the Fourier transform analysis; Banten Province included a monsoon rain pattern with unimodial rainfall. Furthermore, the results of the analysis of the Gumbel method and the Mononobe method, Banten Province included the category of moderate rain and tended to be heavy, even extreme. Based on the results of the analysis using these two methods, in 2025 in Banten Province, it is predicted that 11 % heavy rain, 3 % very heavy rain and 1 % extreme rain are predicted. In that year, it is predicted that there will be 65 sub-districts in Banten Province that have the potential for flooding. The sub-districts that have the potential for flooding are mostly located in Serang Regency, Serang City, Tangerang City, and South Tangerang City. This potential flood is predicted to occur in: January, February, March, April, May, October and November