HIGH 5! Here is the news update of the latest trends from the global technology and startup scene. HIGH 5 is published by Lab1886. This week’s topics are:
Remote-controlled trucks +++ Future of computer vision +++ DIY app with AI object recognition +++ Peer-to-peer: Solar power for EVs +++ Recovering heat from subway tunnels
Autonomous // Freight
Last week, the first automated truck from startup Starsky Robotics drove on the Florida Turnpike, controlled by a remote driver who drove the truck at speeds of up to 55 mph, changing lanes and passing through a tollbooth. The startup hopes to combine highway automation with teleoperation, so that in the future, “truck pilots” can remotely control trucks between distribution centers and highways.
Today, computer vision works with deep learning algorithms that use convolutional neural networks (CNN) to decipher images such as street signs for autonomous vehicles. In the future, image captioning software combined with natural language generation (NLG) could interpret objects for people with visual impairments. Computer vision will play a decisive role in the evolution of AI to human-equivalent levels of intelligence and ultimately into artificial superintelligences superior to humans.
An IBM developer has published a software code pattern which any programmer can use to train AIs to recognize objects. The code pattern, called “Create a real-time object detection app using Watson Machine Learning” is a blueprint that can be integrated into any iOS app, creating a wealth of potential new applications for users.
Source: The Next Web
In addition to feeding surplus electricity from renewable sources into the Smart Grid for electric cars, there is also a local variant. In the Quartierstrom project, researchers at ETH Zurich are seeking to build a local energy market by working with 37 households. Of these, 28 produce their own solar power and can pass it on to all participants at need – not through official clearinghouses, but using a distributed blockchain technology.
Researchers at EPFL’s Soil Mechanics Laboratory (LMS) in Switzerland have developed a geothermal heat recovery system that could transfer heat from over-hot subway tunnels to surrounding households. They have devised a formula for the energy transfer coefficient which could be used to develop a cost-effective, energy-efficient system for pumping heat to the surface and using it to heat – or cool – nearby homes.
Source: New Atlas
HIGH 5 is published by Lab1886. If you have any questions or suggestions, please contact us using the following email address: David Menzel (email@example.com)