Self-driving is currently the Holy Grail in the automotive world, with a number of companies racing to build general-purpose autonomous vehicles that can get from point A to point B with no user input. While no one has brought one to market yet, at least one has promised this feature and had customers pay for it, but continually moved the goalposts for delivery due to how challenging this problem turns out to be. But it doesn’t need to be that hard or expensive to solve, at least in some situations.

The situation in question is driving on a single stretch of highway, and only focuses on steering, so it doesn’t handle the accelerator or brake pedal input. The highway is driven normally, using a webcam to take images of the route and an Arduino to capture data about the steering angle. The idea here is that with enough training the Arduino could eventually steer the car. But first some math needs to happen on the training data since the steering wheel is almost always not turning the car, so the Arduino knows that actual steering events aren’t just statistical anomalies. After the training, the system does a surprisingly good job at “driving” based on this data, and does it on a budget not much larger than laptop, microcontroller, and webcam.

Admittedly, this project was a proof-of-concept to investigate machine learning, neural networks, and other statistical algorithms used in these sorts of systems, and doesn’t actually drive any cars on any roadways. Even the creator says he wouldn’t trust it himself, but that he was pleasantly surprised by the results of such a simple system. It could also be expanded out to handle brake and accelerator pedals with separate neural networks as well. It’s not our first budget-friendly self-driving system, either. This one makes it happen with the enormous computing resources of a single Android smartphone.

It wasn’t long ago that faced with a controller project, you might shop for something with just the right features and try to minimize the cost. These days, if you are just doing a one-off, it might be just as easy to throw commodity hardware at it. After all, a Raspberry Pi costs less than a nice meal and it is more powerful than a full PC would have been not long ago.

When [Joe Coburn] wanted to make a pan and tilt webcam he didn’t try to find a minimal configuration. He just threw a Raspberry Pi in for interfacing to the Internet and an Arduino in to control two RC servo motors. A zip tie holds the servos together and potentially the web cam, too.

You can see the result in the video below. It is a simple matter to set up the camera with the Pi, send some commands to the Arduino and hook up to the Internet.

The serial protocol for the Arduino is simple: The Pi sends a numeric position followed by a P (for pan) or T (for tilt) at 9600 baud. A web server and some Python handle the interface to the Internet and the human.

We’ve certainly seen our share of similar projects. Some of them have been a bit larger.

Primary image

The idea in short:
I will send a robot around planet earth. The robot will be sent to you free of charge. Let it run in your area for 24h and show all earthlings your projects or a piece of your country. Send the rover to the next destination after your mission is over (postal charges will be refunded).
The robot can be controlled through a web interface while transmitting a live video stream. All young scientist and of course all discoverers that are young at heart get free access to the robots, there is even no registration needed.

read more

[Dann Albright] writes about some small experiments he’s done in home security.

He starts with the simplest. Which is to purchase an off the shelf web camera, and hook it up to software built to do the task. The first software he uses is the free, iSpy open source software. This adds basic features like motion detection, time stamping, logging, and an interface. He also explores other commercial options.

Next he delves a bit deeper. He starts by making a simple motion detector. When the Arduino detects motion using a PIR sensor it gets a computer to text an alert. After the tutorial begins to veer a little and he adds his WiFi light bulbs to the mix. Now he can send an email and change the color of the lights.

We suppose, that from a security standpoint. It would really freak a burglar out if all the lights turned red when they walked into a room. Either way, there’s definitely a fun weekend project in playing around with all these systems.

Smart Doorbell is the name of the project for a new tutorial created for Intel Edison. It’s a motion detecting doorbell, that senses when someone is near the door(bell) and via a webcam sends a picture of the person approaching the door to a web address while playing a sound.

“Motion detecting video streaming doorbell“, a medium-advanced level tutorial, is intended to be a good starting point for building basic IoT devices. You’ll learn how to use a webcam to detect motion: when activated, the device will play a greeting, take a photo and email it to someone to let them know who is at the door.
Take a look at the video to see how it works and follow these steps to make yours:

A company called Studio Diip has created a motorized go-cart. For fish. Though "why" is the first question that comes to mind, at this point we're only able to give you the "how." When Little Dory starts to swim in a particular direction, her movement is captured by a top-mounted web cam, processed by a Beagleboard and used to steer the Arduino-controlled carriage under her bowl. As seen in the video after the fold, that lets her follow her muse outside the aquarium's confines. We're hoping this experiment makes it beyond the lab one day, as we're pretty damned curious to see where a fish wants to go. ("Away from the cat" or "towards the ocean" seem like safe bets.)

Filed under: Misc, Transportation

Comments

Source: Studio Diip

The LVL1 Hackerspace held a hackathon back in June and this is one of the projects that was created in that 24-hour period. It’s a 3D scanner made from leftover parts. The image gives you an idea of the math used in the image processing. It shows the angular relations between the laser diode, the subject being scanned, and the webcam doing the scanning.

The webcam is of rather low quality and one way to quickly improve the output would be to replace it with a better one. But because the rules said they had to use only materials from the parts bin it worked out just fine. The other issue that came into play was the there were no LCD monitors available for use in the project. Because of that they decided to make the device controllable over the network. On the right you can see a power supply taped to the top of a car computer. It connects to the laser (pulled out of a barcode scanner which produces a line of red light) and the turntable. A Python script does all of the image processing, assembling each slice of the scan into both an animated GIF and an OBJ file.

[Thanks Nathan]

Hackaday reader [Tom Price] often uses Skype to communicate with family near and far, but he was getting tired of adjusting his webcam each time his kids moved out of frame. While the solution he came up with isn’t fully automated, it is hands-free, which is good enough for his purposes.

[Tom] was looking around for an electronic foot pedal of some sort when he came across a wireless 3rd party Guitar Hero peripheral that happened to fit the bill. Using an Arduino library created by [Bill Porter], he was quickly able to get the toy to communicate with an Arduino-flashed Atmega8, but things kind of fell flat when it came time to relay signals back to his computer. Using another Atmega8 along with the PS2X library, he was able to emulate the Guitar Hero controller that his foot pedal was looking for.

With the pedal portion of his project wrapped up, he focused on his webcam. [Tom] mounted the camera on a small servo, which he then wired up to the receiving end of his foot pedal rig. As you can see in the video below, he can now pan his camera across the room with a tap of his foot, rather than leaning in and manually adjusting it.

Hey,

The ModuPong YouTube channel is about a modular robotics system that tracks a ping pong ball in real-time. In addition, we cover in this channel how the system predicts and responds in less than 130ms.

Please tell me if you have any ideas on how we could make our system better. Cheaper and faster would be great!