Sensors for Smart Robotics & IoT
I. Sensors Overview
Project your mind out to the world of the future. The place where robotics, not just software, has become embedded everywhere. Everything has a smart sensor. Of course, many people refer to this trend as the Internet of Things (IoT), but we don’t like that term because it doesn’t tell you what those things are. In terms of sensors, there are many different types, but vary based on the application
Smart Phones & Wearables
- Heart Rate
II. The Future of Sensors for Robotics
These are just a few of the tens of thousands of different types of sensors in the world, all for different use cases. You can see how combining different types of data with a new kind of intelligence and hardware can give you new forms of world-changing technologies that never existed before.
At the highest level, all of these things are considered robotics. Of course, when it moves into the consumer area, it’s given a new name. Like a smart phone or a self-driving car. But truth be told, the real definition of robotics is just:
Robotics = Sensors + Software
As we move into ever more complex robotics, it requires new forms of high-end sensors and new forms of high-end software.
That’s why a Self Driving Car = LIDAR + Deep Learning.
But tomorrow, the equation will change again:
Humanoid Robot = All Sensors + Biologic Intelligence
Why? Because the sheer complexity of movement and manipulation in a humanoid form means that it’s going to require ever more powerful intelligence that works far more efficiently and with zero previous training data than Deep Learning.
III. Illustration: Smart Tires
There was a report by TechCrunch for Goodyear’s AI Tires which, you guessed it, has a smart sensor embedded within the tire itself, processing the messy data right there at the edge instead of sending it back to a hulking data center. From the article:
"Using data collected by these sensors, the tire could then activate built-in actuators to change the shape of the surface and tread of the tire. It’s a bit like how electrical signals tell your muscles to change shape when you flex or grip, but all done at the behest of on board virtual intelligence telling the tire what shape will best help it maintain traction and control given the current state of the road beneath it."
It still requires the manual human labor process of collecting data, cleaning it, hand-tagging it, creating a deep learning model, being careful not to over-optimize it to the test data you have, then putting it into production to see how it works, then tweaking the model and running through the process again and again.
Take a wild guess where we’re going to point you instead of this archaic AI method.
According to a report from CBS News, almost half of United States citizens are represented in a facial recognition database...
Kaggle has a bit of a history with Google. Earlier in March 2017, Google and Kaggle teamed up to host a $100,000 machine learning competition around classifying YouTube videos...
Asset Optimization is a structured approach to improving the effectiveness of overall asset management by classifying, defining, scheduling and implementing initiatives for better performance and decision making...
Google is also developing another deep learning algorithm that surveys biopsy images to locate metastatic breast cancer that has spread to the lymph nodes. The algorithm was tested...