In this tutorial, Lukas Biewald (a former Stanford Robotics Lab engineer) shows us how he builds two robots able to run deep learning to do object recognition.

Both robots use the open source software library TensorFlow. The library comes with a prebuilt model called “inception” that performs object recognition.

Deep learning and a large public training data set called ImageNet has made an impressive amount of progress toward object recognition. TensorFlow is a well-known framework that makes it very easy to implement deep learning algorithms on a variety of architectures. TensorFlow is especially good at taking advantage of GPUs, which in turn are also very good at running deep learning algorithms.

Object Recognizing Robot from $100 of Parts and TensorFlow

Plum Geek project description:

The Spirit Rover is the next big step in advanced robots for learning, teaching, and all around hacking fun. The robot is outfitted with up to three different computing processors, built with high quality components, and fit into an iconic form factor that any tech nerd can appreciate!

Spirit is a perfect starting point for students and hobbyists looking for an expandable and full featured robot platform. Whether you’re new to coding or involved in serious robotics research, the Spirit Rover has something for you.

• Learn and expand your Python coding knowledge
• Learn and expand your C/C++ Arduino skills
• Learn and apply computer vision
• Design your own autonomous rover missions
• Learn and expand advanced Linux skills

Programmed with Python and Arduino
Want to learn to code in Python and/or Arduino? Whether you’re new to programming or a pro, the capabilities of the Spirit Rover hardware will allow you to grow and apply your skills. Many combinations of programming are possible. Write your code using Python and C/C++ on the Raspberry Pi, or write your code in C/C++ using the free and open source Arduino environment. Our easy to use functions allow seamless communication between the two boards.

Three Computer Boards in One Robot
The Spirit Rover robot includes three different computers, just like many other advanced robots you’ll find in the real world. You’ll learn how these more advanced systems really work at the low level.

A Raspberry Pi computer will handle most of your processing. Though it is optional, it is a powerful computer capable of doing many things at one time. The Pi is similar to the computer inside a tablet computer or small laptop.

An Arduino compatible processor can be used alone or together with the Pi. This is the same processor as found on the popular Arduino UNO board. It is also the same processor (and runs the same code!) as the processor on our Ringo, Wink, and Plumduino boards.

A Microchip PIC processor handles the low level processing on the robot. It does things like sending pulse signals to the servos, reading light sensors, and managing the power system. It is pre-loaded with code. Normally you won’t play with this code on your own, but it is still open and hackable if you want to customize it.

Spirit Rover – Learn Raspberry Pi and Arduino the fun way!

This rover looks like one ready to take Elon Musk on a journey to Mars, but no. It’s a prototype build by Steelsquid, the same guys that released a minimal and optimized version of Raspbian called Steelsquid Kiss OS.

The project is impressive and all the details are here:
Squidsscout FPV rover

This is an autonomous unmanned ground rover (UGV) driven by four hacked servos Turnigy S8166M with 33kg/cm2 torque and 48rpm at no load.

The rover is able to deal with a variety of terrains from sand to mud and grass.

The designer shows the process of building this UGV on his blog. Read the details here.

Ryan Zotti:

I’ve spent the past 6 months building a self-driving toy car using a Raspberry Pi, OpenCV, and TensorFlow. If you’ve ever thought about building your own self-driving toy car, this presentation will help you avoid common pitfalls and shed light on important tradeoffs that you’ll have to weigh along the way. I’ll cover things like how to parse images, how to effectively tune machine learning neural

We celebrate the new Raspberry Pi board with three fresh projects from robotics. The new Pi Zero is definitely a hit in the hobbyists area since the first 20,000 boards were sold out in the first 24 hours. This information is impressive even for a well-known prototyping board from the Pi family.

After a few days of Zero life, we discover some interesting projects in robotics. But this is just the beginning. We’re still waiting to see more distinguishes and innovative ideas to work with this Linux board.

Here are the first three projects with Pi Zero:

1. PiZero controlling a PiBorg 4Borg robot
   
2. Raspberry Pi Zero controlling Cam Jam EduKit from ThePiHut
   
3. Raspberry_Pi trying out the Pi Zero in our humanoid robot the HROS1
   

This self-driving car project is a piece of DIY artificial intelligence designed by Zheng Wang. The project is the perfect choice for those who wants to deal with self-driving rovers on roads.

The designer modifies a simple RC car to handle three tasks:

1. self-driving on the track;
2. stop sign and traffic light detection;
3. front collision avoidance;

All these three tasks have been possible due to three subsystems:

1. input unit with a camera and an ultrasonic sensor;
2. processing unit with a Raspberry Pi;
3. RC car control unit;

Check the links below for the guide and all the code needed to make this self-driving robot car.

OpenCV Python Neural Network Autonomous RC Car | GitHub
Self Driving RC Car | zhengludwig.wordpress

Free professional robotics projects always come in handy. Therefore, in most cases, it’s useful to think of free robot projects as a playground for your imagination. Whether you are looking for inspiration, professional coding or engineering solutions — in all three cases you can learn a lot, you can apply them, and you can build customized robots upon them without reinventing the wheel all the time.

A large robot rover is hard to design, require more time and, even more, a considerable budget. However, to build a unique robot rover, sometimes it’s not enough to assemble some parts or write some lines of code.

In this article, we present a project designed by Andreas Nilsson.

The Large Rover is a 50x50cm robot with 20cm wheels and controlled using an Internet connection and a webpage. The rover features a Raspberry Pi camera to enjoy real-time video streaming.

Key Pieces:

• PiBorg Diable DC motor controller
• Raspberry Pi camera
• 4 DC motors
• A servo motor
• Steelsquid PIIO board
• Raspberry Pi
• Wifi adapter
• Steelsquid Kiss OS

Check the link below for the guide and all the code needed to build from scratch this large robot rover.

Large Rover | Steelsquid

Even More DIY Robots!

Here are more articles with references to over 30 DIY robots we’ve managed to collect so far.

• 25+ Fresh and Impressive DIY Robots
• Print Your Own Kame Quadruped Robot
• Insanely DIY Self-Driving Robot Car

Last But Not Least

We are regularly looking for the most innovative DIY robot projects. If you build one of them, please contact us at office@intorobotics.com — we would like to support you on Into Robotics.

Raspberry Pi is usually the first option to consider when we talk about single board computers in robotics and automation. However, this development board is focused on supporting a variety of operating systems including Linux distributions and the Windows 10 IoT Core. For instance, if you plan to run the Windows 10 IoT Core, the operating system is not always intuitive, and you need good support for an efficient use of the operating system.

However, you don’t need to search through thousands Windows 10 features and extension to solve every Raspberry Pi problem. You can search through professional tutorials which provide excellent information, and of course, can save you a lot of time.

The Raspberry Pi 2 and Raspberry Pi 3 are our two pieces of exciting projects with Windows 10. Despite the Raspberry Pi 2, the Raspberry Pi 3 has built-in Wireless and Bluetooth LE, which make easier to connect the board to the Internet. In addition, the Windows 10 IoT Core adds support for the official Raspberry Pi Wi-Fi dongle and other common Wi-Fi dongles.

Eleven months ago we had presented a hand-picked collection of Raspberry Pi 2 OS selection for designers and developers. Now it’s time for a fresh list of Windows 10 learning tutorials. This article provides professional Windows 10 tutorials aiming to help you get Windows IoT Core skills quickly and improve the quality of your applications.

Setup

• In this Instructables tutorial, you’ll find how to install Windows 10 on your Raspberry Pi 2. All these steps are also valid for the new Raspberry Pi 3.
• This is how to manually update the Windows 10 IoT Core.
• Do you know how to connect to your Raspberry Pi 2 or 3 using the SSH network protocol? This tutorial reveals the steps to connect at your Pi through several operating systems including Linux, Windows, and Android.
• If you use C# as your main programming language, this is how to make a connection between Raspberry Pi 2 with Windows 10 IoT to a PC.

How To IoT

• Even if you’re using the Raspberry Pi 2 or Raspberry Pi 3 with sensors, this tutorial show you how to use the FEZ HAT board with Windows 10 running on the development board.
• This is how to measure the temperature using the Raspberry Pi 2 and Windows 10 IoT Core.
• Do you know how to read analog data through ADC over SPI on Raspberry Pi? This tutorial shows you how to use the Microchip MCP3008 to read analog data from two proximity sensors.
• These are the steps and code to control remotely an electric device using the Raspberry Pi board and Windows 10 IoT Core.
• Build your first “Hello World!” application with Windows 10 and Raspberry Pi.
• This tutorial shows you how to build an application to display a countdown for your next bus.
• This is how to measure the light intensity in your room and output data to a web API.

How To

• If you want to build mobile and desktop web applications, the Angular framework is a good start. In this tutorial, John Deutscher shows you how to build an application for Windows 10 IoT Core.
• Do you want to build your own voice-enabled coffee maker? This is how to build them.
• This is how to build a plotter with Windows 10 IoT Core and Raspberry Pi.

(Image credits: Github, Johndeutscher, Everydaylinuxuser, Dotnetcurry)

This is a guest post. Thank you Fabrice R. Noreils.

1. The Objective

The main objective is to design a medium size humanoid robot (around 90 cm) and create a niche market which is considering humanoid robot like an art. The robot will be able to walk like a human, thanks to the innovative design of mechanical structure and associated algorithms, perform task and, very important point, equipped with outifts/outshells created by famous designers in order to meet different communities’ expectations.

2. How to reach it

In order to reach this ambitious target, several objectives have been defined:

• The first objective is to create innovative walking gaits – closest as possible to human ones, i.e. no more bending knees – real heel strike – active use of an articulated forefoot. These gaits will be more or less “hardcoded“. The main goals are in one hand to demonstrate the feasibility of such gaits and on the other hand to understand how the different parts of the body are coordinated and how to take benefit from the dynamic of the robot.
• The second objective is to design a controller, based on the knowledge gained from the previous work, which will be connected to sensors and takes into account the dynamic of the robot. It means that the walking gaits will not be hardcoded anymore but generated online. It will lead to faster and smoother walking gaits as well as the possibility to react to unexpected events/obstacles.
• Once the mobility is achieved, it can be possible to add more sensing capabilities like a camera, different kinds of gripping tools and/or specific accessories in order to achieve new tasks.
• The last objective is more oriented towards art and design. To please the audience, mobility and instilled the audience that the robot is “more or less clever” is not enough, it must be beautiful. To achieve this goal, the idea is to bring famous designers that will be able to create outfits/outshells matching different kind of communities’ expectations (fashion, mangas, mecha…).

3. Results

The first objective of this project is under completion and in order to show the progress, mainly focusing on the walking gaits, I published videos that are available on YouTube.

Two different gaits are considered: straight walk gait and Turning gait. For each one there are several videos (Video 1 is the oldest, video n is the newest).

odoi Straight Walk Video 2 with speedx4

Odoi straight walk Video 3 speedx4

The first video regarding the turning gait:

More videos will be added from time to time and technical details can be found on the blog I have created: Artbot.

4. The robot

This section provides a detailed description of both the mechanical structure as well as the hardware of the actual robot.

Figure 1 gives an overview of the current robot. It is 75cm tall.

4.1 Mechanical structure

The robot is equipped with

• An articulated feet;
• An articulated pelvic;
• An articulated torso;

Most of the brackets are made of resin (and printed with a Form 1+) except the Pelvic where aluminum has been required. Indeed because of the robot weight, the previous brackets and Pelvic structure made of resin bent leading to discrepancies between real angles and theoretical ones at the hip that were too important.

The foot which is an important element of the robot is detailed on Figure 2.

The main benefits of the mechanical design are:

• Innovative gait – no more bending knees;
• Walking gait closer to human gait;
• Omni-directional walking;
• Save energy as the robot is not bending knees which is a real benefit when the robot runs on
  batteries;
• Possibility to change the stride length;
• Participation of the whole body.

So far, from my knowledge only two hobbyists/researchers were able to achieve a walking gait without bending knees:

• Masahiko Yamaguchi, nickname is Dr Guero, featured a modified KHR-3HV which was able to walk on a floor almost like a human does [1]. However the robot is not able to turn and I do not know if it is possible to change the stride length and get the same astonishing result.
• Tomotaka Takahashi with his latest creations, Robi, Kirobo/Mirata and more recently Robohon, developed a patented walking gait without bending knees [2] – However these cute robots are equipped with a quite large footprint in order to maintain stability.

4.2 Hardware

The hardware – see Figure 3– will be composed of:

• An OPEN CM9 board which is connected to the Dynamixel servos from Robotis. The robot is equipped with 2 * MX106T, 4*MX64T, 11*MX28T and 9 AX12 servos;
• A PIXY Cam to do basic object/color recognition;
• A 9DOF Razor IMU which will be used to control the balance;
• FSR sensors;
• Murata Rotary potentiometers to measure the heel orientation;
• A Raspberry Pi 2/3 will be added soon in order to generate gaits (and more) online.

In the current version of the robot, only the OPEN CM9 is used and it is connected to the Dynamixel servos. It will be connected to the FSR and Murata sensors very soon.

The Raspberry will be added in the coming months. It will be then connected to the OpenCM9, the Pixy CAM and the 9DOF Razor IMU.

5. How to do

The principle of the approach is depicted on Figure 4.

Each walking gaits is decomposed into phases:

• Lateralization on the right side;
• Left leg swing;
• Lateralization on the left side;
• Right leg swing.

For each phase, a sequence of movements is created. Each sequence is composed of a start (T0) a duration (D), an objective () and a template curve to follow.

The phases are the input of the kinematic simulator which generates the movements of each limbs. This software takes into account the kinematics of the robot, it computes also the CoG (Center of Gravity). The output is a set of angular positions at different timestamps for each limb.

In order to compute the angular position and associated speed for each servo, another software is used, referred to as the “commands generator”.

The inputs are the angular position for each limb (processed by the Kinematic simulator) and the duration of each phases. The possibility to change the duration of each phase gives us the possibility to play (a bit) with the dynamics. If the duration is too short, the velocity at a given time for some actuators will be too important and it will generate oscillations leading to instabilities.

The output of the “commands generator” is a table of commands that will be used by the OpenCM9 board. If a list of commands have to be sent at a given time t, then a pointer to a list of (servoId, position and speed) is created – see Figure 5.

6. Artistic design

I think that Artistic Design is really very important if one want to introduce robots in the human environment that can be accepted and/or tolerated by the population. One step further will be the development of “artistic robots” that can be considered as piece of Art. So far only the Japanese robotics community is really addressing this topic.

This is why I am eager to work with designers in order to create “outer shell” (or even outfits) that can fit the robot skeleton. I initiated a collaboration with Dacosta Bayley, a Canadian artist, who is running among other things MarchOfRobot on Instagram (every day in March, artists are pushing sketches picturing robots – see #marchofrobots2016). He ran successfully a kickstarter campaign in 2014 in order to publish a book about his work.

7. Next steps

The next steps will consists in bringing all the software on the raspberry in order to generate “on the fly” the walking gaits. Sensors will be connected to the processing boards in order to adjust the gaits to the reality of the terrain. Dynamics will be introduced step by step.

Carry on a project like this one alone is very challenge, therefore if there are some people interesting to participate with skills in software programming, mathematics, electronics or artistic, please do not hesitate to contact me.

References

1. https://www.youtube.com/watch?v=-Vg-BdXps50
2. https://www.youtube.com/watch?v=RPjlJKRa7Uw

Have you ever wondered how fast and far a Raspberry Pi robot runs throughout the day and night? With GPS capabilities and a Linux computer such as Raspberry Pi, you can track a robot or a multirotor and locate these on a map. But in general speaking, with a Raspberry Pi and a GPS unit you can build limitlessness indoor and outdoor applications.

In this article, I explored the most popular GPS systems compatible with Raspberry Pi boards and used in the DIY area. In the list are included GPS add-ons, expansion boards, dongles, shields and other modules such as RasPiGNSS, EM-506, NEO-7M-C, BU-353, the Adafruit Ultimate GPS, 3DR uBlox, and Navigatron v2 – I2C.

If some GPS units cost as much as four cups of coffee at Starbucks, there are also less accessible modules that exceed the price of 100.00€ ($113.00).

In this collection, I present you 14 GPS units with support for Raspberry Pi Model A+, B or B+. So fire up Raspberry Pi and get ready to get your hands dirty.

01. GPS add-on

The GPS add-on compatible with Raspberry Pi B and B+

• How to setup the add-on: Raspberry Pi + ITead Studio GPS;

02. RasPiGNSS

The RasPiGNSS expansion board

• Installation guide here;

03. GPS expansion board

Another tracking expansion board for Pi

On top of the board can be attached a battery to keep on the settings in the event of power loss.

04. USB GPS Dongle

The USB GPS Dongle

• A tutorial to start working with the USB electronic device: Getting GPS to work on a Raspberry Pi;

05. GPS shield

The stackable GPS shield for Pi

• Step-by-step tutorial to start using the shield: GPS Module for Raspberry Pi Tutorial;

06. EM-506

The €35.00($40) GPS module is another receiver based on the SiRF StarIII chipset. Like the USB GPS dongle described above, the EM-506 provides the position very accurate even in urban canyon and dense foliage environment. The features include a position accuracy of 2.5m, and without any network assistance, it can predict for up to three days the satellite positions.

07. NEO-7M-C

With 56 receive channels and an IPX interface, the NEO-7M-C is easy to use and has a price of €24.00 ($28.00). The receiver is engineered to support a large variety of software like Google Earth, u-center and more.

08. Dexter Industries GPS

The stackable Dexter Industries GPS

• An instructables tutorial that shows you how to setup and start receiving the data from the Dexter Industries shield: GPS and the Raspberry Pi;

09. BU-353

The BU-353 USB dongle

Designed to work with any Linux computer, the €29.00 ($33.00) USB GPS receiver has a high sensitivity and an accurate position of 10 meters.

• Short guide to install the BU-353 on a Raspberry Pi: Raspberry BU-353 installation;

10. Adafruit Ultimate GPS

The Adafruit Ultimate GPS system

With a position accuracy of 1.8 meters and a velocity of 0.1 m/s, the Adafruit GPS Breakout is a very sensitive device for high speed movements. It has a price of €35.00($40.00) and a power consumption of only 20 mA during navigation.

• A guide to connect the module to Pi board: Adafruit Ultimate GPS on a Raspberry Pi;

11. 3DR uBlox

The 3DR uBlox compatible with with 3DR APM 2.6 autopilot system

Designed for multicopters and rovers, the €79.00($90.00) GPS module is based on the HMC5883L digital compass. And because it’s a system for flying robots, the uBlox supports configuration to work with 3DR APM 2.6 autopilot system.

• Tutorial to connect the uBlox with Pi: How to connect 3DR GPS uBlox to Raspberry Pi;

12. GSM/GPRS & GPS shield

The GSM/GPRS & GPS shield

• Step-by-step tutorial that shows you how to add GSM, GPRS and GPS functionality to Raspberry Pi: A GSM/GPRS & GPS Expansion Shield for Raspberry Pi;

13. 3G/GPRS shield

A 3G and GPRS shield for Pi

The 3G/GPRS shield is a device designed for Internet of Things applications. And because we are talking here about GPS data, the shield also provides the location and stay connected to the 3G network. The price is huge, about €149.00 ($158.00), and it’s compatible with Pi, Intel Galileo and Arduino boards.

• Here is a tutorial to start working with the 3G/GPRS shield: New 3G + GPS shield for Raspberry Pi tutorial;

14. Navigatron v2 – I2C

The Navigatron v2 – I2C navigation system

• Guide to use the Navigatron with Raspberry Pi: Making an autonomous boat using a Raspberry Pi;

The Raspberry Pi camera guide can be broken down into 8 dimensions:

1. determine what you need;
2. determine if you already own any potentially compatible camera;
3. define the automatic functions of your camera;
4. the compatibility between camera and your Raspberry Pi model;
5. how to interface the camera to the prototyping board;
6. accessories for cameras;
7. what type of camera has the best performance;
8. documentation to setup and build applications;

All these eight key factors should help you figure out which is the best digital camera for your DIY project.

The Raspberry Pi camera guide: best practices

1. Determine what you need

You should be demanding when looking to buy a camera. After all, you together with the Pi and this little gadget are going to do many things, from capturing hundreds of photos to detect objects or build a point and shoot camera.

To avoid getting sucked into buying a camera beyond what you really need, here are a few questions to ask yourself before shopping:

What do you need the camera for?
Whether you choose the cheapest or the most expensive camera to use it with the Pi board, there are plenty of applications to build.

Any of these tiny cameras allows you to build a stream live feed, doing macro filming and photography, reading a barcode, process images captured by a robot, detect the motion, build a face detection and recognition application, and why not build your homemade point and shoot camera.

What conditions will you largely use the camera?
Do you plan to use the camera in low or bright light? Indoor or outdoor? The Pi is a prototyping board focused on robotics, automation and aerial applications. So you can use this little computer anywhere and in different conditions.

Night vision captures
If you plan to build an application to take a capture in night vision conditions, you need a special camera with infrared capabilities.

What is your budget?
Are you interested in purchasing a digital camera and you’re not sure which is the best one for your budget?

With cameras ranging from $6.00 to $69.00, the decision boils down to different specifications, work without additional accessories such as a power hub and price.

2. Do you already own any potentially compatible digital camera?

If you don’t use until now the Pi for video and image captures, is unlikely to have in the drawers any specially designed camera for Raspberry Pi.

Rather than using a dedicated camera module for this tiny prototyping board, you can use a standard USB webcam to capture images and videos on the Raspberry Pi memory. Here is a long list with USB webcams compatible with Raspberry Pi.

3. Automatic functions

Understanding the automatic modes of a camera can make a real difference to the quality of your images. But from these eight cameras explored in this article, only three of these are featured with automatic functions: Raspberry Pi NoIR, Raspberry Pi Camera Module and Logitech Webcam C525.

The Raspberry Pi Camera Module and NoIR shares the same automatic image control functions including the automatic exposure control (AEC), automatic white balance (AWB), automatic band filter (ABF), automatic 50/60 Hz luminance detection and automatic black level calibration (ABLC).

The Logitech webcam C525 is featured with an ultra-smooth autofocus and auto light correction for dim and harsh lighting.

4. Is the digital camera compatible with your Raspberry Pi model?

Before digging deep into what makes the cameras a compatible product, I’ll show you the Raspberry Pi family tree:

1. Raspberry Pi B
2. Raspberry Pi A
3. Raspberry Pi B+
4. Raspberry Pi A+
5. Raspberry Pi 2 Model B

If there is no doubt that the Raspberry Pi Camera Module and SainSmart Infrared Night Vision Camera are compatible with all Pi versions (B, A, B, A, 2 Model B), the Pi NoIR Camera Board is compatible only with the first versions of the Pi: Model A and Model B.

As you might expect, all the USB webcams are compatible with any Raspberry Pi model, but these four media gadgets were tested on the Raspbian and Debian Wheezy operating systems with versions no older than 2014. So, in the worst case if you run an old version of Raspbian or Debian Wheezy, you have to update your operating system with an up-to-date version.

5. Interface

Using the CSI connector to interface a camera module has advantages and disadvantages. To install a camera via the CSI interface is very simple and allows you to capture images in minutes. Simply plug the ribbon cable into the CSI interfaces and setup the camera using the standard Pi program.

With only one CSI connector on the board, the biggest disadvantage is that the Raspberry Pi can host only one digital camera. The Pi NoIR, SainSmart surveillance camera, and the official hardware add-on for the Raspberry Pi use the dedicated CSI connector.

The Pixy camera is different and support SPI, I2C, UART, USB or analog/digital output.

But if you were able to use one of the general-purpose USB-connected cameras, the Pi can handle any number of USB cameras.

6. Accessory

Every time when Raspberry Pi introduces a new model, we all faced with a series of questions about compatibility. The Raspberry Pi camera it’s itself an accessory for the prototyping board and this accessory needs other accessories.

If you work outdoors, you need a protective camera case and a wall mountable support that holds the Pi and digital camera firmly in place.

Another great accessory is an adjustable mount bracket designed to hold the camera in multiple positions.

You can have a much wider perspective of the world if you take a step back and view the Raspberry Pi images through a wide-angle lens. The Pi camera is engineered with a fixed focus lens and a wide-angle camera lens increase the viewing angle of the camera.

A Raspberry Pi or a NoIR camera module comes with a 15 centimeters flat cable. Depending on the project specifications, sometimes you have to move the camera where you want. Because the fruit named computer is so small, this should not be a problem. However, in case that this small and flexible cable is not enough, you can use a long flexible cable to increase the distance between Pi and camera.

If all these accessories are designed for dedicated Raspberry Pi cameras, the list of accessories for the USB webcams is very short, almost doesn’t exist.

7. Performance

When working with robotics and you need high performances to run computer vision applications, you need GPU and CPU performances.

In this case, we have two camera types: camera board or USB camera. Because the Pi camera board connects directly to the GPU, it has only a little impact over the CPU performance, leaving it available for other processes and applications. In other words, the dedicated cameras are faster and provide a higher resolution than any other USB cameras.

8. Documentation

Once you have the hardware components, you need documentation to put these together. In other words, you need documentation to setup the camera and capture images/videos.

All dedicated cameras that use the CSI connector comes with good online documentation to interface and build applications with your camera.

On the other side, the USB cameras or Pixy comes with good documentation for their core applications, but almost nothing in the DIY area. In this case, you have to find the necessary software and commands to turn any USB webcam into a digital eye for the Raspberry Pi.

The Raspberry Pi camera list

Different features are the essence of all these cameras. In today’s world of infinite gadgets and smartphones that can be integrated with a Raspberry Pi computer, you must understand what makes these cameras different and why you have to use one of these and not any other digital camera. In the following, I’ll tell you the key characteristics for all digital cameras presented in this guide.

The Raspberry Pi camera specifications

Because the world of cameras that encodes digital images and videos digitally with the Raspberry Pi computer is small compared with compact cameras or DSLRs, in just a few words I’ll do a short description of the eight best digital cameras compatible with the Raspberry Pi board.

1. Raspberry PI 5MP Camera Board Module
The Raspberry PI 5MP camera module	This camera is for general work. It should work for anything as long as the light is required. With a weight of just over 3g, the camera is perfect for applications where the size and weight are important such as drones or tiny mobile robots. The 5 megapixels sensor is capable to support 1080p30, 720p60 and 640x480p60/90 video and a maximum resolution of 2592 x 1944 pixels for static images.
2. Raspberry Pi NoIR Camera Module
The official Raspberry Pi NoIR camera	This is the official night vision camera for Raspberry Pi designed without the infrared filter on it. This missing feature makes it perfect to take infrared images in low light conditions.The camera has a native 5MP sensor that can deliver 1080p30, 720p60 and 640x480p60/90 video recording and 2592 × 1944 pixels for static images.
3. SainSmart Infrared Night Vision Camera
A good camera for night vision applications	This is a special camera with an infrared night vision system for surveillance and any other application in low light conditions. The camera has a native 5MP sensor capable of 2592×1944 pixels for static images and 1080p@30 fps, 720p@60fps or 640x480p 60/90 in video recording.
4. Pixy CMUcam5
This is the Pixy CMUcam5	This camera is fast enough to detect and track objects around it, but the object should be colored in bright solid color to make a good distinction from the surroundings. The most common way to interface the camera module with a Pi is via the Pixy’s UART or I2C lines. The image sensor is capable of 1280×800 pixels for static images and 720p HD resolution for video.
5. Sony Playstation Eye for PS3
You can play with your Pi and the Sony Eye for PS3	Tested on Wheezy and Raspbian operating system, the Sony Playstation Eye for PS3 can deliver only 640×480 or 320×240 pixels for video and static images. The camera doesn’t require a power hub.
6. Logitech Webcam C525
One of the best USB camera	The Webcam C525 is equipped with the best sensor of all digital cameras with USB and deliver an HD 720p resolution for video capture or 1920×1080 pixels for static images.
7. HP Webcam HD-2300
The HP webcam for your Pi project	Like the MiniCam Pro, the Webcam HD-2300 was tested on the Raspbian and Wheezy operating system and deliver an HD 720p resolution at 30 fps or 1280×720 pixels resolution for static images.
8. GE MiniCam Pro
A cheap USB camera	The MiniCam Pro is a cheap USB camera that can deliver a range of resolution between 640×480, 352×288, 320×240, 176×144 and 160×120 pixels. The camera was tested on a Pi running Raspbian or Wheezy operating system and can run without a powered hub.

If you have a Raspberry Pi 2, what do you use it for? It seems a light question, but the answer could take hundreds of lines.

This fruit-named computer is a precious tool for students, artists, and of course hobbyists and hackers. With features for developing things in different areas, it is not a surprise to have hundreds of thousands if not millions of users.

With this credit card-sized development board, you can build robots for $50, $100, $1000, $10.000 … and more. You can make fun applications and robots just to play in your house, or you can try to build amazing things that are supposed to push forward the Raspberry Pi limits.

To create complex robots, you do not need to re-invent the wheel. In fact, you can combine a large variety of existing solutions, technologies and concepts.

However, if you are planning on designing a Raspberry Pi 2 robot, these are the most creative projects that you can reach right now. Yes, you have to see these project briefs pushing your creativity to inspire your next project.

01. Pi Tank

The PiTank design is a mix of a card-sized computer for prototyping things and the 3D printing concept.

Thanks to PiTank designers, this project is open-source and easy to replicate by anyone at home.

The purpose of this robot is to create an intelligent “pet” for any maker or hacker able to write software for it. There are three main components: a Kinect sensor, a smartphone used to control the platform, and the Raspberry Pi 2 board that controls the components of the robot.

If you love to build tracked robots, this robot is one of the cheapest projects that you can build at home from scratch.

02. Autonomous rover with vision

This autonomous robot is a complex project that requires more than a Raspberry Pi 2 board. Only a Pi 2 is not enough to control all the systems integrated into the platform.

The design of the robot consists of 4 essential processes. The first process includes the data from sensors. The second process manages the guidance of the machine. The others two processes include the time sensitive control systems while the final routine handles the speaker.

Given all of these features, the Raspberry Pi 2 board has an essential role. The Pi board is the machine vision system of this robot.

If you want to integrate the Pi 2 with microcontrollers, add a wide range of sensors and a large amount of data, this project is perfect for such a complex combination.

03. Terrestial Explorer V1.0

This is an all-terrain rover that features sensors to scan the environment and wireless equipment to transmit data to the operator.

Like many other complex robots, this one combines the Arduino microcontroller with the power of Raspberry Pi 2. An XBee module is attached to the Arduino microcontroller and sends data from sensors to a remote control with an LCD screen attached.

With sensors able to measure the methane gas levels, temperature, humidity, light intensity, UV, etc., this rover is a good project that connects the Arduino microcontroller with the 900MHz quad-core ARM Cortex-A7 CPU of the Raspberry Pi 2.

04. Ballin’ Octo Robot

Most of the time, this is a remote controlled truck with machine vision capabilities. The robot truck has a camera attached on top of it and captures images for color tracking and chasing a ball.

This project combines the Arduino microcontroller and the Raspberry Pi 2 board. The Pi board captures video images to send commands to the Arduino and video images to the user.

In autonomous mode, the robot uses the camera captures to detect and follow a ball.

Given all these features, this project includes methods for processing, analyzing, and understanding images with the Raspberry Pi 2.

05. Segway with Raspberry Pi

This wheeled balancing robot is a special project that uses the Pi 2 to control the DC motors, read the 6-axis motion sensor data, run a PID algorithm together with a program that ensures the protection of the robot.

In this project, the designer chooses the Raspbian as the operating system. The Raspbian OS takes the advantage of a good support in the Raspberry Pi community as a respectable Linux distribution.

The Pi board is programmed to measure the orientation of the robot and drive the motors 100 times per second. To protect the robot body, if the angle of orientation is larger than 60 degrees, the motors are stopped as a safety measure.

06. The HoloLens robot

This robot is a special project that uses the Raspberry Pi 2 board with Windows 10 OS and the HoloLens platform for augmented reality. The HoloLens is a head-mounted computer that brings to live the Raspberry Pi 2 robot with a body and a control panel.

This mix between holographic projections and how a robot would behave change the virtual exploration as we know and use it and how we build robots.

07. Roberta

Roberta is another self-balancing robot, but this time much simpler to build. The brain of the robot is a Pi 2 board controlled via remote control.

08. Raspberry Pi 2-wheel balancing robot

It seems that the 2-wheeled balancing robot is a popular application for Pi 2 makers. This project combines LEGO parts with the Raspberry Pi 2 features.

09. Humanoid Robot Controlled by a Raspberry Pi

It is expensive to build a humanoid robot. It’s like making an investment into robotics. An open-source framework such as OpenCv cut the costs of the whole project. And this robot uses the Pi board to run OpenCV applications for autonomous skills.

10. DiddyBorg

What can you do with a Playstation 3 remote control and a 6 wheeled robot platform? Well, there are a lot of options, but this robot kit combines in the best way the 6 axis Playstation 3 remote control and the power of a Raspberry Pi 2 board.

11. Raspberry Pi 2 drone

This Raspberry Pi 2 drone uses the NAVIO+ autopilot shield to determine the position of the drone and control the actuators. This is a solution to cut the costs instead investing money to buy pricey control systems for your drone. Combining the Raspberry Pi features with the NAVIO+ capabilities of sensing the world, is a cheap and friendly solution.

Engineers have always tried to give the robot the gift of sight. So, they have to replicate the human vision process with computers, algorithms, cameras and more.

In the DIY area, a Raspberry Pi is the queen of prototyping platforms. It’s useful in different areas and for a large variety of applications. So, why not to use it in computer vision applications. The projects started coming fast and furious for navigation, localization, recognition, classifications, monitoring, reading and more.

There is virtually no limit what can be done with a single board computer, a camera module, a vision library such as OpenCV and a bit of creativity.

As you will see from the tutorials explored in this article, some of the most popular applications in computer vision deals with the detection, tracking and the recognition of objects and humans. Whether you are looking to build a robot able to detect a human or an automated system able to detect an object, the Raspberry Pi board is the center of your project.

From this collection of hand-picked tutorials, you will learn all kinds of tricks that can be applied to build simple and cost effective computer vision applications based on Pi.

Navigation and Obstacles Avoidance

• Navigation to a target
  On the bigfacerobotics, Peter Neal shows us in a tutorial how to build an autonomous robot able to navigate to a target by detecting the coloured border of an image.
  
• Programming a Raspberry Pi Robot Using Python and OpenCV
  In this project, the designer looking to make an autonomous robot with the py_websockets_bot library. The Python library communicates with the mobile robot over a network interface and sends commands that control the movements of the robot.
  
• The RR.O.P. – RaspRobot OpenCV Project
  This Raspberry Pi robot uses the shapes, colors and textures of the objects to interact with the external environment.
  
• Final Project Car Lab
  In this project, the designers build a computer vision application to avoid obstacles on a wide path defined by black parallel lines.
• OpenCV and python for a line follower
  With a webcam, the OpenCV library, Python and a Raspberry Pi board, you can build a line follower robot using computer vision algorithms.
  
• Obstacle detection using OpenCV
  In this tutorial, the designer uses four steps to detect obstacles in front of the robot. The first step is to capture an image. The second step is to convert the image into a grayscale image. The third step is to blur it slightly, and in the fourth step uses canny edge detection to highlight the edges in the image.
  
• Autonomous bottle recycling robot
  In this tutorial, you can find how to build an eco-friendly robot engineered to avoid obstacle until the camera detects and recognizes a bottle.
  

Tracking and Recognition

• Wall E Robot (object/sound recognition, AI)
  This tutorial is inspired by the Wall-E robot from the movie and shows you how to detect and track an object.
  
• OpenCV&Pi Cam – Step 7 : Face recognition
  An example how to use the Pi board with OpenCV to detect human faces.
  
• Raspberry Pi Motion Following Network Camera Project
  If you plan to build a motion tracking network pan and tilt camera, this tutorial shows you how to build it from scratch.
  
• Basic motion detection and tracking with Python and OpenCV
  This is an example how to build a vision system for home surveillance. The system is designed to detect and track human faces.
  
• Raspberry Pi and the Camera Pi module: face recognition tutorial
  In this tutorial for a facial recognition system, Boris Landoni shows you how to locate human parts like faces, eyes, noses, and more.
  
• Raspberry Pi Color Tracking Using PID
  Colors are everywhere. In this tutorial, is explained how to use an OpenCV PID control algorithm to track objects.
  
• Using a Raspberry Pi, Arduino and Python for disguise detection
  The idea of this project is to build a DIY vision system to replace the thermal cameras. In this guide is described how to use multispectral imaging by combining a high-resolution visible wavelength camera with a low-resolution IR sensor.
  
• Object Tracking on the Raspberry Pi with C++, OpenCV, and cvBlob
  This project was built to track objects, and the algorithm is implemented in both OpenCV and cvBlob.
  
• Footfall: A Camera Based People Counting System for under £60
  This tutorial shows you how to build a cheap vision system to count people that visit a building or are located in the building.
  
• Real time Drone object tracking using Python and OpenCV
  This is a simple project to track objects from the air with a flying drone.
  
• Tutorial: Using CamShift to Track Objects in Video
  This is another tutorial from Adrian Rosebrock that shows you how to use the CamShift algorithm to find and track objects in a video.
  

Object Sorting

• Scanning woes revisited: a Raspberry Pi scanning machine
  You can build a vision application able to detect if a page should be or not in a document. This tutorial explains to you how the vision system works.
  

  OLYMPUS DIGITAL CAMERA

• 3 Ways to Compare Histograms using OpenCV and Python
  With OpenCV and Python, you have three options to compare histograms. All these three options are described in this tutorial.
  
• Raspberry Pi / Lego Ball Machine
  Do you like to build robots with LEGO blocks? In this tutorial, you can find how to build a ball sorter machine with LEGO blocks, Raspberry Pi, a camera module and OpenCV.
  

You may be interested in the following Raspberry Pi resources as well:
The Raspberry Pi camera guide

If you want to be among the first who play with the new board from Raspberry Pi, below are the online stores from where you can order a Raspberry Pi Zero W.

The letter W stands for Wireless. The first small plate from Raspberry Pi is called Zero and doesn’t have integrated WiFi or Bluetooth. The difference between Zero and Zero W is the Wireless and Bluetooth modules that are integrated into the board.

The price remains very attractive. Most online stores have a price that starts at $10. At that price, you have to add the accessories, so that the final price may be several times higher.

• thepihut

  thepihut

• modmypi

  modmypi

• canakit

  canakit

• kubii

  kubii

• kiwi-electronics

  kiwi-electronics

• samm

  samm

• pimoroni

  pimoroni

• adafruit

  adafruit

The English company Freeburn Robotics Limited, which includes PiBorg, has launched a new campaign to raised funds on Kickstarter. The raising funds campaign has exceeded the target of £ 3,000 from the first day of release. The amount is not very big, but it is still impressive how quickly they managed to attract funds needed to start the production. The project is a good idea for the hobbyists who are using Raspberry Pi to control robots.

PiBorg is specialized in products compatible with Raspberry Pi. They have released in the recent years a wide range of mobile platforms and accessories compatible with all the Raspberry Pi boards. Among the popular products is the platform DiddyBorg with six wheels.

Returning to the current campaign, PiBorg has designed and will launch on the market two new products: a mobile robot with four wheels and a motor controller.

MonsterBorg and ThunderBorg

The four-wheel mobile platform is MonsterBorg. Few details about the subject:

• doesn’t require soldering or the use of complex tools. You buy the platform, and with a screwdriver and some effort, you have an intelligent robot with Raspberry Pi;
• the mobile platform is compatible with Raspberry Pi 3, Pi 2, B+ and Pi Zero. Given that the kit will not contain a Raspberry Pi board, it is your choice what version of Raspberry Pi you will use to control the robot;
• the chassis is made of aluminum, so resistant to scratches, etc.;
• the wheels have a diameter of 105mm / 4 inches. So you can play with the robot in the park, garden, or you can get dirty in the mud in the woods;
• the motor controller can output up to 5A per channel. So you will have four powerful DC motors to drive the platform;
• you can mount a lot of accessories compatible with the Raspberry Pi board. From cameras to servo motors and sensors, to anything else that can be controlled by Pi;

The motor controller is ThunderBorg. Few details about it:

• as I said above, the controller can output up to 5A on each of the two channels;
• the supply voltage is between 7V and 36V. In other words, you can use the controller to control a wide range of DC motors;
• it has PWM;
• it has over-current protection, short circuit protection and over temperature protection;

You can find the Kickstarter campaign here.

The recipe for Niryo One is as follows: 3D printing, Raspberry Pi 3, Arduino Mega, RAMPS 1.4, ROS (Robot Operating System), Linux Ubuntu for Raspberry Pi, and lots of open-source code.

Let us study each feature:

3D printed:
All the components of the robotic arm that can be printed, have been printed with a 3D printer. The producers have used PLA as the printing material, but other materials may also be used.

Using the 3D printing technology to build most parts of the robot, the final price of such a project is lower when compared to traditional methods to build the same parts of a robot. Another benefit is that you can print components at home, or replace them if necessary.

Raspberry Pi 3: WiFi, Bluetooth, Ubuntu, ROS, Python.
Pi 3 connects the robot arm to the Internet or to a mobile device via WiFi and Bluetooth. Also, Pi runs important programs to control the Arduino board, and programs written in Python. In other words, Pi 3 running all programs that cannot run on the Arduino Mega.

Arduino Mega: the RAMPS 1.4 shield, control of DC motors, control of sensors.
That’s what Arduino does in this project. Read data from the sensors and control the DC motors. The data and commands are flowing through the RAMPS 1.4 shield.

RAMPS 1.4
RAMPS is a shield specifically designed to be compatible with the Arduino Mega board. This shield can control up to 5 stepper motors and few servo motors. It is interesting that such shields are used to build 3D printers. So, if you want to reuse some of the components of the robotic arm, you can build a 3D printer.

ROS: algorithms, applications
ROS running on the Raspberry Pi 3. The framework is designed to let the user add intelligence to the robotic arm. How? For example, it can add a camera and write an application for processing and analyze the images. In other words, the robotic arm can be programmed to recognize objects and sort them by color, size, etc. Moreover, ROS is open source and has a very active community.

Programs: open-source, GitHub
All the programs developed for Niryo will be available on GitHub. These programs can be downloaded and used to control the robotic arm.

Niryo One

LOCORO is an open-source project available to roboticists enthusiastic to work with Raspberry Pi, ROS, and Linux. Here, I would be adding the parts that can be printed at home with a 3D printer. In conclusion, the final dimensions of the robot may differ depending on the requirements and needs.

Let’s go back to the interesting part, the smart components. What should be noted here is:

• the robot brain Raspberry Pi 3 runs Raspbian. Pi 3 control sensors, motors, and almost everything must be controlled
• ROS does what it does best. Allows the addition of capabilities such as mapping or computer vision
• electronic component assembly is here
• the instructions for hardware assembly is here
• here are the steps for software
• the web application can be found here
• the program written in Python is here
• and the parts that can be printed can be found here, including the wheels

LOCORO