Category robots in business

Building a custom training loop in Tensorflow and Python with checkpoints and Tensorboards visualizations

Robots will help H-E-B grocery stores keep up with the growing demand for online grocery services amid the pandemic.

Robots will help H-E-B grocery stores keep up with the growing demand for online grocery services amid the pandemic.

In the current issue of Science Robotics, researchers from Istituto Italiano di Tecnologia (IIT- Italian Institute of Technology) and Centro Protesi INAIL in Italy reported on their ability to replicate the key biological properties of the human hand: natural synergistic and adaptable movement, biomimetic levels of force and speed, high anthropomorphism and grasp robustness. Developed by a collaborative of researchers, orthopaedists, industrial designers and patients, the prostetic hand called Hannes is able to restore over 90% of functionality to people with upper-limb amputations.

Skoltech researchers have designed and developed an interface that allows a user to direct a small drone to light-paint patterns or letters through hand gestures. The new interface, DroneLight, can be used in distant communications, entertainment, and even search and rescue. The paper was published on the preprint server arXiv.org and presented at IEEE International Conference on Robot & Human Interactive Communication.

• Mobile robots will take logistics sector by storm, with a 10-fold investment increase predicted between now and 2024 • COVID-19 will be a significant driver in long-term mobile robot growth • Sales of AMRs set to radically outstrip AGV revenues

A sentin VISION system uses AI-based recognition software and can be trained using a few sample images. Together with a GigE Vision CMOS industrial camera from IDS and an evaluation unit, it can be easily embedded in existing processes.

In this episode, Audrow Nash interviews Eric Diller, Assistant Professor at the University of Toronto, on wireless micro-scale robots that could eventually be used in human surgery.  Diller speaks about the design, control, and manufacture of micro-scale surgical robotic devices, as well as when we might see this technology in the operating room.

Eric Diller

Links

• Download mp3 (12.2 MB)
• Subscribe to Robohub using iTunes
• Subscribe to Robohub using Spotify
• Subscribe to Robohub using RSS
• Support us on Patreon

Mars, also known as the red planet, has been the focus of numerous research studies, as some of its characteristics have sparked discussions about its possible inhabitability. The National Aeronautics and Space Administration (NASA) and a few other space agencies have thus sent a number of rovers and other spacecraft to Mars with the hope of better understanding its geology and environment.

As for the TS5000, its communication functions enable real-time data transmission between the robot and controller, allowing manufacturers to integrate the system into a smart factory environment.

1. Transfer Learning

Transfer learning is used a lot in machine learning now since the benefits are big. The general idea is simple. You train a big neural network for purposes with a lot of data and a lot of training. When you then have a specific problem you sort of “cut the end off” the big network and train a few new layers with your own data. The big network already understands a lot of general patterns that you with transfer learning don’t have to teach the network this time. 

A good example is if you try to train a network to recognize images of different dog species. Without transfer learning you need a lot of data, maybe a 100.000 images of different dog species since the network has to learn everything from scratch. If you train a new model with transfer learning you might only need 50 images of every species.

You can read more about Transfer Learning here.

2. Active learning

Active learning is a data collection strategy that enables you to pick the data that your AI models would benefit the most from when training. Let’s stick with the dog species example. You have trained a model that can differentiate between different species but for some reason the model always has trouble identifying the german shepherds. With an active learning strategy you would automatically or at least with an established process pick out these images and send them for labelling. 

I made a longer post about how active learning works here.

3. Better data

I’ve put in a strategy here that might sound obvious but is sometimes overlooked. With better quality data you often need way less data since the AI does not have to train through the same amount of noise and wrong signals. In the media AI is often talked about as “with a lot of data you can do anything”. But in many cases making an extra effort to get rid of bad data and make sure that only correctly labeled data is used for training, makes more sense than going for more data.

4. GAN’s

GAN’s or Generative Adversarial Networks is a way to build neural networks that sounds almost futuristic in it’s design. Basically this kind of neural network is built by having two networks compete against each other in a game where one network creates new fake training data examples from the data set and the other is trying to guess what is fake and what is real data. The network building fake data is called the generator and the network trying to guess what is fake and what is real is called the discriminator. This is a deep learning approach and both networks keep improving during the game. When the generator is so good at generating fake data that the discriminator is consistently having problems separating fake from real we have a finished model. 

For GAN’s you still need a lot of data but you don’t need as labelled data and since it’s usually the labelling that is the costly part you can save time and on your data with this approach.

5. Probabilistic Programming

One of my very favorite technologies. Probabilistic programming has a lot of benefits and one of them is that you can often get away with using less data. The reason is simply that you build “priors” into your models. That means that you can code your domain knowledge into the model and let data take it from there. In more many other machine learning approaches everything has to be learned by the model from scratch no matter how obvious it is. 

A good example here is document data capture models. In many cases the data we are looking for is obvious by the keyword to the left of it. Like “ID number: #number# is a common format. With probabilistic programming you can tell the model before training that you expect the data to be to the right of the keyword. Many neural networks are taught from scratch requiring more data.

You can also read more about probabilistic programming here: https://www.danrose.ai/blog/63qy8s3vwq8p9mogsblddlti4yojon

Japanese grocery chain FamilyMart has teamed up with Tokyo startup Telexistence to test the idea of using a remotely controlled shelf stocking robot named the Model-T to restock grocery shelves. On its website, Telexistence describes the robot as a means for addressing labor shortages in Japan and also as a way to improve social distancing during the pandemic.

Researchers at University of Colorado Boulder's ATLAS Institute and University of Calgary have recently developed an actuated, expandable structure that can be used to fabricate shape-changing aerial robots. In a paper pre-published on arXiv, they introduced a new robot, dubbed PufferBot, which was built using this unique and innovative structure.

Researchers at University of Colorado Boulder's ATLAS Institute and University of Calgary have recently developed an actuated, expandable structure that can be used to fabricate shape-changing aerial robots. In a paper pre-published on arXiv, they introduced a new robot, dubbed PufferBot, which was built using this unique and innovative structure.

George Konidaris still remembers his disheartening introduction to robotics.