Skip navigation
For my last part of this road test I'm looking at the stall guard function. This works by sensing the current through the motor coils so my first step was to look at the datasheet chapter 9 on sense resistors. Thanks to jancumps for reviewing the schematic earlier in the roadtest and identifying that the evaluation board uses a 0.06Ω resistor. Note that there are two sense resistors, one for each of the motor coils. The motor current can then be determined with the following calculation ...
The next test for the TMC5161 was positional accuracy. This is where my test rig came in. I fixed the motor to one bracket and attached the 3D printed drum. A waxed cotton thread was wrapped around the drum twice for grip and then fed around the idler at the far end of the test rig. The tension was provided with some rubber bands. Finally a card was added in the middle of the rig and the cord was marked with white paint. This time the position mode in the TMCL-IDE was used along with the pos ...
To test the accuracy of the stepper driver I am using a large test rig with a pointer to indicate the position. There is also an optical encoder mounted on the back of the stepper motor and the TMC5161 chip can read the pulses from this and make it's own conclusions about missed steps. So thought it would be good to wire up the encoder and see how it behaves.   Wiring the encoder  From the datasheet ABN Incremental Encoder Interface The TMC5161 is equipped with an incremental encoder ...
The next neural network that I'm going to try is a variant of Tiny-YOLO.  The You Only Look Once (YOLO) architecture was developed to create a one step process for detection and classification.  The image is divided into a fixed grid of uniform cells and bounding boxes are predicted and classified within each cell.  This architecture enables faster object detection and has been applied to streaming video.   The network topology is shown below.  The pink colored layers h ...
Before I move on to object detection I thought I would try one more example of object classification using a more complex neural network based on the Multi-layer offload architecture.  The network used is a variant of the DoReFa-Net and uses the large ImageNet dataset http://www.image-net.org/  for training.  The DoReFa-Net https://arxiv.org/pdf/1606.06160 is a low bitwidth convolutional neural network that is trained with low bitwidth gradients optimized for implementation on har ...
After my previous blog post it was pointed out to me that the amount of whitespace (or other non-object pixels, i.e. background) that I had in my captured image was affecting the accuracy of the classification.  I did a quick inverse test that was suggested by beacon_dave and added whitespace around the CIFAR-10 test image and it then classified as an airplane!  I guess this makes sense as there are a lot of extraneous pixels to confuse the classifier.  In general purpose use of a ...
In the previous blog PYNQ-Z2 Dev Kit - CIFAR-10 Convolutional Neural Network , I verified the 3 hardware classifiers against the reference "deer" test image.  Now I'm going to see how the classifiers perform with captured webcam images.  I expect the performance will be degraded because the webcam will produce lower quality images due to issues like image brightness and focus.  CIFAR-10 has a small training set (5000 images per class), so I'm going to use a solid background to hel ...
The first neural network implementation that I'm going to look at is for CIFAR-10 (Canadian Institute For Advanced Research).  CIFAR-10 is a computer vision dataset used to train and test neural networks for object recognition.  The CIFAR-10 data consists of 60,000 32x32 color images in 10 classes, with 6000 images per class. There are 50,000 training images and 10,000 test images.   Labeled Image Classes airplane automobile bird cat deer dog frog horse ship truck   ...
Introduction What is a Data Acquisition System? Market Review Cost of Ownership Module Types Build Quality Power Consumption Tests Performance Tests Operation with the BenchVue software. Summary Introduction  This is my opening blog on the DAQ970A data acquisition unit that I was lucky enough to be selected to road test. This is a general blog, with my intention to introduce the DAQ970A.   I didn't really do an unboxing as they don't do an awfu ...
Introduction  As Particle introduced their third generation of hardware. The Particle Argon, Boron, and Xenon are Particle’s latest offering in the world of IoT dev boards, and this time they add mesh networking. The three new boards are all built around the Nordic nRF52840 SoC and include an ARM Cortex-M4F with 1MB of Flash and 256k of RAM. This chip supports Bluetooth 5 and NFC. Breaking the new lineup down further, the Argon adds WiFi with an ESP32 from Espressif, the Boron brings ...
One of the technical areas that I am interested in is object detection in video streams.  The specific application is the real-time identification of objects in video from IP surveillance cameras.  As part of the PYNQ-Z2 roadtest I want to see how well an FPGA implementation of a neural network works for this task.  A key to an efficient implementation (power and area) in programmable logic is quantization.  I came across an interesting paper on quantization while researching ...

Filter Blog

By date: By tag: