ai

340 Articles

AI Makes Linux Do What You Mean, Not What You Say

April 20, 2021 by 30 Comments

We are always envious of the Star Trek Enterprise computers. You can just sort of ask them a hazy question and they will — usually — figure out what you want. Even the automatic doors seemed to know the difference between someone walking into a turbolift versus someone being thrown into the door during a fight. [River] decided to try his new API keys for the private beta of an AI service to generate Linux commands based on a description. How does it work? Watch the video below and find out.

Some examples work fairly well. In response to “email the Rickroll video to Jeff Bezos,” the system produced a curl command and an e-mail to what we assume is the right place. “Find all files in the current directory bigger than 1 GB” works, too.

Continue reading “AI Makes Linux Do What You Mean, Not What You Say” →

Death Of The Turing Test In An Age Of Successful AIs

April 6, 2021 by 98 Comments

IBM has come up with an automatic debating system called Project Debater that researches a topic, presents an argument, listens to a human rebuttal and formulates its own rebuttal. But does it pass the Turing test? Or does the Turing test matter anymore?

The Turing test was first introduced in 1950, often cited as year-one for AI research. It asks, “Can machines think?”. Today we’re more interested in machines that can intelligently make restaurant recommendations, drive our car along the tedious highway to and from work, or identify the surprising looking flower we just stumbled upon. These all fit the definition of AI as a machine that can perform a task normally requiring the intelligence of a human. Though as you’ll see below, Turing’s test wasn’t even for intelligence or even for thinking, but rather to determine a test subject’s sex.

Continue reading “Death Of The Turing Test In An Age Of Successful AIs” →

An AI-Free Way To Catch Wildlife On Camera

March 29, 2021 by 14 Comments

Judging by the over-representation of the term “AI” in our news feeds these days, we’re clearly in the exponential phase of the artificial intelligence hype cycle, and very nearly at the dreaded “Peak of Inflated Expectations.” It seems like there’s nothing that AI can’t do, and nowhere that its principles can’t be applied to virtuous — and profitable — effect.

We don’t deny that AI has massive potential, but we strongly suspect that there will soon come a day when eyes will roll and stomachs will turn at yet another AI application that could have been addressed with something easier. An example of the simpler approach can be seen in this non-AI wildlife photo trap, cobbled together by [Sebastian] to capture pictures of some camera-shy squirrels. Rather than train an AI with gigabytes of squirrel images, he instead relies on his old Sony Alpha camera, which has a built-in WiFi. A Python script connects to the camera, which is trained on a feeder box and set to a very narrow depth of field. That makes a good percentage of the scene out of focus until a squirrel or other animal comes along looking for treats. The script detects the increased area of the scene that is now in-focus with a Laplace operator in OpenCV, and triggers the camera shutter. [Sebastian] ended up with some wonderful shots of the shy squirrels using this scheme; the video below describes the setup in more detail.

It’s not the first time we’ve seen Laplace transforms used to gauge image sharpness, of course, but we really like the approach [Sebastian] took here for its simplicity. The squirrels are cute too.

Continue reading “An AI-Free Way To Catch Wildlife On Camera” →

An ESP Will Read Your Meter For You

February 7, 2021 by 17 Comments

As home automation starts to live up to its glossy sci-fi promise there remains a deficiency when it comes to interfacing between the newer computerised components and legacy items from a previous age. A frequent example that appears in projects on Hackaday is the reading of utility meters, and in that arena [jomjol] has a very neat solution involving an ESP32 camera module and a software neural network to identify meter readings directly.

The ESP and camera sit at the top of a 3D-printed housing that fits over the meter. The clever trick comes as each photo’s orientation is determined, and not only is OCR used to read digits but also figures are derived from small dial meters and other indicators on the meter face. It’s a very well-thought-out system, with a web-based configuration tool that allows full customisation of the readable zones and how they should be treated.

This project makes full use of the ESP32’s capabilities, and the attention to detail that has gone into making it usable is particularly impressive. It certainly raises the bar against previous OCR meter reading projects.

[Thanks for the tip Sascha]

AI Learns To Drive Trackmania

January 17, 2021 by 26 Comments

Machine learning has long been a topic of interest for humanity, but only in recent years have we had broad access to great computing power to enable to the average person to dive in. [Yosh] recently decided to put an AI to work learning how to race in Trackmania.

After early experiments with supervised learning, [Yosh] decided to implement a genetic algorithm to produce an AI to drive in the game. The AI takes distance from the track walls as an input, and has steering and accelerator values as an output. Starting with 100 AIs in generation 1, [Yosh] iterated by choosing the AIs that covered the longest distance in 13 seconds. Once the AIs started to get the hang of the first few corners, he changed the training to instead prioritize the lowest time taken to traverse each of the checkpoints along the track.

The AI improved over time, and over 100 generations, got down to a 23.48s time on the test track, versus 19.63s for [Trabadia], a talented human. We’d love to see how much better the AI could do with more training. [Yosh] is trying more experiments, like providing extra feedback in the AI fitness function to keep it from hitting the walls. It’s not the first time we’ve seen a genetic algorithm used to train a racing AI, either. Video after the break.

Continue reading “AI Learns To Drive Trackmania” →

Baby Yoda Becomes Personable Robot

January 12, 2021 by 6 Comments

Baby Yoda has been a hit character in Disney’s The Mandalorian, but does not actually exist in real life as far as we know. Instead, [Manuel Ahumada] set about building a robotic replica, complete with artificial intelligence.  (Video, embedded below.)

The first step was to build a basic robotic simulcra of Baby Yoda, which [Manuel] achieved by outfitting a toy with servos, motors and a Raspberry Pi. With everything hooked up, Baby Yoda was able to move his head and arms, and scoot around on wheels, all under the control of a Bluetooth gamepad. With that sorted, [Manuel] added brains in the form of a smartphone running Intel’s OpenBot machine learning platform. This allows Baby Yoda to track and follow people it sees on its smartphone camera, and potentially even navigate real-world spaces with future upgrades.

It’s a fun build, and we’d love to see the bot let loose at a convention to explore and make friends. We’ve covered OpenBot before, and look forward to seeing it used in more builds. Video after the break.

Continue reading “Baby Yoda Becomes Personable Robot” →

AI On The Highway

January 5, 2021 by 13 Comments

A couple of announcements caught our attention last week regarding AI-controlled cars. South Korea’s Kakao Mobility and local startup Autonomous A2G launched a limited self-driving taxi service in Sejong City this month, made possible by enabling legislation passed in May. For now, the service is restricted to government employees, and the AI driver will be backed-up by an engineer who is there to monitor the systems and take over in an emergency. The companies plan to expand the fleet and service areas this year, although no details are given.

Another announcement comes from the Ministry of Land, Infrastructure and Transport about the on-going successes of the semi-autonomous truck platooning program. This is a collaboration between the Korean Expressway Corporation, Kookmin University in Seoul, and Hyundai Motors. Previously restricted to a designated test road called the Yeoju Smart Highway, the program is now being tested on public roads at speeds up to 70 kph. This year the program will expand to platoons of 4 trucks running at 90 kph. We’ve always thought that long-haul trucking and freight industries would be an early adaptor AI technologies, and one which AI could offer significant benefits.

Continue reading “AI On The Highway” →