Prostheses are a great help to those who have lost limbs, or who never had them in the first place. Over the past few decades there has been a great deal of research done to make these essential devices more useful, creating prostheses that are capable of movement and more accurately recreating the functions of human body parts. At Georgia Tech, they’re working on just that, with the help of AI.
[Jason Barnes] lost his arm in a work accident, which prevented him from playing the piano the way he used to. The researchers at Georgia Tech worked with him, eventually producing a prosthetic arm that, unlike most, actually has individual finger control. This is achieved through the use of an ultrasound probe, which is used to detect muscle movements elsewhere on his body, with enough detail to allow the control of individual fingers. This is done through a TensorFlow-based neural network which analyses the ultrasound data to determine which finger the user is trying to move. The use of ultrasound was the major breakthrough which made this possible; previous projects have often relied on electromyogram sensors to read muscle impulses but these lack the resolution required.
The prosthesis is nicknamed the “Skywalker arm”, after its similarities to the prostheses seen in the Star Wars films. It’s not [Jason]’s first advanced prosthetic, either – Georgia Tech has also equipped him with an advanced drumming prosthesis. This allows him to use two sticks with a single arm, the second stick using advanced AI routines to drum along with the music in the room.
It’s great to see music being used as a driver to create high-performance prosthetics and push the state of the art forward. We’re sure [Jason] enjoys performing with the new hardware, too. But perhaps you’d like to try something similar, even though you’ve got two hands already? Try this on for size.
Google has announced their soon to be available Vision Kit, their next easy to assemble Artificial Intelligence Yourself (AIY) product. You’ll have to provide your own Raspberry Pi Zero W but that’s okay since what makes this special is Google’s VisionBonnet board that they do provide, basically a low power neural network accelerator board running TensorFlow.
The VisionBonnet is built around the Intel® Movidius™ Myriad 2 (aka MA2450) vision processing unit (VPU) chip. See the video below for an overview of this chip, but what it allows is the rapid processing of compute-intensive neural networks. We don’t think you’d use it for training the neural nets, just for doing the inference, or in human terms, for making use of the trained neural nets. It may be worth getting the kit for this board alone to use in your own hacks. An alternative is to get Modivius’s Neural Compute Stick, which has the same chip on a USB stick for around $80, not quite double the Vision Kit’s $45 price tag.
The Vision Kit isn’t out yet so we can’t be certain of the details, but based on the hardware it looks like you’ll point the camera at something, press a button and it will speak. We’ve seen this before with this talking object recognizer on a Pi 3 (full disclosure, it was made by yours truly) but without the hardware acceleration, a single object recognition took around 10 seconds. In the vision kit we expect the recognition will be in real-time. So the Vision Kit may be much more dynamic than that. And in case it wasn’t clear, a key feature is that nothing is done on the cloud here, all processing is local.
The kit comes with three different applications: an object recognition one that can recognize up to 1000 different classes of objects, another that recognizes faces and their expressions, and a third that detects people, cats, and dogs. While you can get up to a lot of mischief with just that, you can run your own neural networks too. If you need a refresher on TensorFlow then check out our introduction. And be sure to check out the Myriad 2 VPU video below the break.
Around the Hackaday secret bunker, we’ve been talking quite a bit about machine learning and neural networks. There’s been a lot of renewed interest in the topic recently because of the success of TensorFlow. If you are adept at Python and remember your high school algebra, you might enjoy [Oliver Holloway’s] tutorial on getting started with Tensorflow in Python.
[Oliver] gives links on how to do the setup with notes on Python versions. Then he shows some basic setup operations. From there, he has the software “learn” how to classify random points that either fall into a circle or don’t. Granted, this is easy enough to do with traditional programming, so it isn’t a great practical example, but it is illustrative for learning purposes.
Given that it is easy to algorithmically decide which points are in the circle and which are not, it is simple to develop training data. It is also easy to look at the result and see how close it is to the actual circle. You’ll see that it takes a lot of slow learning before the result space looks like a circle and not a triangle or some other odd shape.
One way to run a compute-intensive neural network on a hack has been to put a decent laptop onboard. But wouldn’t it be great if you could go smaller and cheaper by using a phone instead? If your neural network was written using Google’s TensorFlow framework then you’ve had the option of using TensorFlow Mobile, but it doesn’t use any of the phone’s accelerated hardware, and so it might not have been fast enough.
Google has just released a new solution, the developer preview of TensofFlow Lite for iOS and Android and announced plans to support Raspberry Pi 3. On Android, the bottom layer is the Android Neural Networks API which makes use of the phone’s DSP, GPU and/or any other specialized hardware to speed up computations. Failing that, it falls back on the CPU.
Currently, fewer operators are supported than with TensforFlor Mobile, but more will be added. (Most of what you do in TensorFlow is done through operators, or ops. See our introduction to TensorFlow article if you need a refresher on how TensorFlow works.) The Lite version is intended to be the successor to Mobile. As with Mobile, you’d only do inference on the device. That means you’d train the neural network elsewhere, perhaps on a GPU-rich desktop or on a GPU farm over the network, and then make use of the trained network on your device.
What are we envisioning here? How about replacing the MacBook Pro on the self-driving RC cars we’ve talked about with a much smaller, lighter and less power-hungry Android phone? The phone even has a camera and an IMU built-in, though you’d need a way to talk to the rest of the hardware in lieu of GPIO.
You can try out TensorFlow Lite fairly easily by going to their GitHub and downloading a pre-built binary. We suspect that’s what was done to produce the first of the demonstration videos below.
We’ve talked about TensorFlow before — Google’s deep learning library. Crunching all that data is the province of big computers, not embedded systems, right? Not so fast. [Neil-Tan] and others have been working on uTensor, an implementation that runs on boards that support Mbed-OS 5.6 or higher.
Mbed of course is the embedded framework for ARM, and uTensor requires at least 256K of RAM on the chip and an SD card less than (that’s right; less than) 32 GB. If your board of choice doesn’t already have an SD card slot, you’ll need to add one.
[Roland]’s workflow consists of breaking up a recording from his backyard into one second clips, loading them in to a Python program and running some machine learning code to determine whether the clip is a recording of a bat or not and using this to determine the number of bats flying around. He uses several Python libraries to do this including Tensorflow and LibROSA.
The Python code breaks each one second clip into twenty-two parts. For each part, he determines the max, min, mean, standard deviation, and max-min of the sample – if multiple parts of the signal have certain features (such as a high standard deviation), then the software has detected a bat call. Armed with this, [Roland] turned his head to the machine learning so that he could offload the work of detecting the bats. Again, he turned to Python and the Keras library.
We keep seeing more and more Tensor Flow neural network projects. We also keep seeing more and more things running in the browser. You don’t have to be Mr. Spock to see this one coming. TensorFire runs neural networks in the browser and claims that WebGL allows it to run as quickly as it would on the user’s desktop computer. The main page is a demo that stylizes images, but if you want more detail you’ll probably want to visit the project page, instead. You might also enjoy the video from one of the creators, [Kevin Kwok], below.
TensorFire has two parts: a low-level language for writing massively parallel WebGL shaders that operate on 4D tensors and a high-level library for importing models from Keras or TensorFlow. The authors claim it will work on any GPU and–in some cases–will be actually faster than running native TensorFlow.