1,721,013 research outputs found
A novel approach to 3D measurement: a system with variable focus lens and a depth from defocus algorithm
No full textDEFINITION OF A MEASURABILITY THRESHOLD OF GEOMETRIC TOLERANCES IN RELATION TO MEASUREMENT UNCERTAINTY AND DIMENSIONAL PARAMETERS
No full textIn order to avoid final product malfunction and to allow
for assembly integration, geometric specifications and
dimensional tolerances are commonly used in mechanical
design. However the feasibility of geometric specification
measurement and verification is often neglected and the
influence of measurement uncertainty in geometric
tolerances evaluation underestimated. The authors propose
updated results of a mathematical and numerical model,
based on Monte Carlo simulations, developed in order to
define a measurability threshold of geometric tolerances in
relation to measurement uncertainty and geometric
parameters, such as feature dimensions, meant to help the
designer to define measurable geometric specifications.
Starting from EN ISO 14253-2:2011 and EN ISO 14253-
3:2011 standards, a perpendicularity tolerance between a
cylindrical feature and a planar one has been simulated. A
mathematical model has been defined for each feature, in
order to assess both misalignment and its uncertainty when
starting from the estimate of geometric entities obtained
from point coordinates measured by a Coordinate
Measurement Machine (CMM).
Monte Carlo Analysis of these simulation underlined
how geometric parameters, such as dimensions of the
features involved, can act as magnifiers for measurement
uncertainty when verifying a geometric specification: there
could be cases where this magnification effect could lead to
non-measurability of misalignment and non-verifiability of
the geometrical specification requested
Deep learning-based hand gesture recognition for collaborative robots
Full text linkThis paper is a first step towards a smart hand gesture recognition set up for Collaborative Robots using a Faster R-CNN Object Detector to find the accurate position of the hands in RGB images. In this work, a gesture is defined as a combination of two hands, where one is an anchor and the other codes the command for the robot. Other spatial requirements are used to improve the performances of the model and filter out the incorrect predictions made by the detector. As a first step, we used only four gestures
Healthcare Sensor System Exploiting Instrumented Crutches for Force Measurement during Assisted Gait of Exoskeleton Users
No full textPowered exoskeletons can be used by the persons with complete spinal cord injury to achieve bipedal locomotion again. The training required before being able to efficiently operate these orthotics, however, is currently based on the subjective assessments of the patient performance by his therapist, without any quantitative information about the internal loads or assistance level. To solve this issue, a sensor system was developed, combining the traditional gait analysis systems, such as ground reaction force platforms and motion capture systems, with Lofstrand crutches instrumented by the authors. To each crutch three strain-gauge bridges were applied, to measure both axial and shear forces, as well as conditioning circuits with transmission modules and a triaxial accelerometer. An inverse dynamics analysis, on a simplified biomechanical model of the patient wearing the exoskeleton, is proposed by the authors as a tool to assess both the internal forces acting on shoulders, elbow, and neck of the patient, as well as the loads acting on joints. The same analysis was also used to quantify the assistance provided to the patient during walking, in terms of vertical forces applied by the therapist to the exoskeleton. The tests showed a therapist assistance contribution reported as a fraction of the subject body weight up to 40% with an average close to 0% and a standard deviation value of 14%. This paper presents the description of the measurement system, of the post-processing analysis, as well as the results of the proposed approach applied to a single Rewalk user during training. © 2016 IEEE
First Step Towards Embedded Vision System for Pruning Wood Estimation
Full text linkThis paper focuses on the development and evaluation of a portable vision-based acquisition device for vineyards, equipped with a GPU-accelerated processing unit. The device is designed to perform in-field image acquisitions with high-resolution and dense information. It includes three vision systems: the Intel® RealSenseTM depth camera D435i, the Intel® RealSenseTM tracking camera T265, and a Basler RGB DART camera. The device is powered by an Nvidia Jetson Nano processing board for both simultaneous data acquisition and real-time processing. The paper presents two specific tasks for which the acquisition device can be useful: wood volume estimation and early bud counting. Acquisition campaigns were conducted in a commercial vineyard in Italy, capturing images of vine shoots and buds using the prototype device. The wood volume estimation software is based on image processing techniques, achieving an RMSE of 2.1 cm3 and a mean deviation of 1.8 cm3. The buds detection task is obtained by fine-tuning the YOLOv8 model on a purposely acquired custom dataset, achieving a promising F1-Score of 0.79
Techniques for on-board vibrational passenger comfort monitoring in public transport
Full text linkTraffic calming devices on urban streets, such as elevated pedestrian crossings, speed bumps and roundabouts, are increasingly used, raising a real problem in relation to the on-board comfort that passengers perceive. To measure vibrational comfort related to traffic calming devices that passengers of the public transport perceive, an acquisition system called ASGCM (Autonomous System for Geo-referenced Comfort Measurements) has been developed, taking as a reference the European regulations on rail transports. ASGCM permits to link each measurement of vibration, ground velocity and acceleration with geographical information resulting from a GPS. In this way a map of a comfort index, statistical surveys and correlation between on-board comfort and traffic calming, can be directly obtained by any Geographic Information System (GIS), able to query a centralized remote database, which was developed ad- hoc. A large number of experimental tests has been performed to define a vibrational comfort index and to collect a large dataset that allows statistically significant comparisons between different infrastructures and their characterization. The proposed technique can also be useful for diagnostics purposes, such as vehicle comparison and road maintenance state monitoring
STEWIE: eSTimating grapE berries number and radius from images using a Weakly supervIsed nEural network
Full text linkCounting tasks with overlapping and occluded tar-gets are often tackled by means of neural networks outputting density maps. While this approach has been proven to be highly effective for crowd-counting tasks, it has not been exploited extensively in other fields (like fruit counting). Furthermore, this approach has never been used to infer the shape or the size of the recognized objects. In this paper, we present a novel deep learning-based methodology to automatically estimate the number of grape berries present in an image and evaluate their average radius as a double output of the network. For the model training, we employ a public dataset consisting of 300 vines images, where each berry center has been dot-annotated. Since the dataset does not directly provide information about the berry radii, we first develop a numerical optimization methodology to calculate the radius of the berries, by exploiting the dot annotations, some prior knowledge (berry maximum size), and a current state-of-the-art segmentation model. Then, we employ the combined information (berry center and radius) to train a custom neural network that outputs two density maps, from which we infer the number of berries in the image and their average size
MIMO NON-LINEAR SENSORS CALIBRATION BASED ON GENETIC ALGORITHMS
No full textThe increasing integration between electronics and
mechanical engineering brings to the industrial market very
hi-tech sensors, often non-linear, capable of more than a
single input and single output. A problem more and more
relevant for sensors like these is calibration. Classic l
calibration procedures, when applied to this extremely
engineered sensors, lead to poor accuracy and are generally
not satisfactory.
The case study is the calibration of a bi
laser based position sensor, in particular a positive sensitive
detector, that is an optical position transducer based on
series of photodiodes commonly used as multidimensional
sensor. To perform the calibration a micrometric positioning
table was used to test the whole photodiode active area in
both directions. The sensor studied showed a very linear
behaviour in the central region of the working range, and a
limited nonlinearity closer to the range limits and was to be
used to verify robot movement capabilities; to reduce
uncertainty associated with nonlinearities,
standard, non-linear, calibrations were performed, pointing
out residual values in order to compare different algorithms.
In a previous work, authors have already tested a linear
model against an algorithm based on radial basis functions
(RBF) and Nelder-Mead simplex method. Object of this
paper is the definition of a procedure based on RBF and
genetic algorithms for multi-dimensional interpolation of
data cloud and a comparison between this updated procedure
results and the ones of the previous studied algorithms.
The reference model for calibration was a black box with
two inputs, X and Y position of the laser spot, and two
outputs, voltages Vx and Vy, while the calibration
procedure was split in two separate layers, one for each
output depending on both inputs. Given N data points in a
M-dimensional environment and N values that represent the
non linearity residual, purpose of the algorithm is to
approximate a data cloud with a real function, that is
represented as a sum of a polynomial (linea
radial basis functions, each associated with a different center
(node) and weighted by an appropriate coefficient, that the
procedure also allow to assess.
When no starting guess for nodes are given in input,
nodes coordinates are the output of a non
based on a genetic algorithm, whose goal is to locally
minimize the objective function. The algorithm stops itself
whenever it reaches a certain tolerance level, a user
specified number of nodes or when the previous iteration
has a better value of the objective function. This study has
been performed for various RBF classes, and shows an
increased accuracy, thus a better metrological behaviour,
with respect to the standard linear (planar) calibration model
traditionally used
- …
