Object Tracking Software for Robotics

The human visual system produces an informational revolution in the brain, which processes all the information received and can distinguish any objects in sight. In other words, the eyes see but the brain does all the work for image processing. The brain is the most reliable and the most powerful tool that controls everything, the human visual system is so powerful that we cannot yet speak of an artificial system that can even remotely compare with it. Robots need a system as powerful as the human system and researchers need inspiration to create algorithms that can detect objects, track and estimate their next position. If for a human distinguishing objects is routine and does not require a conscious effort of analyzing the information, for a robot detection with high precision of an object is a task that requires high processing power, an algorithm that identifies objects and a real time image capture system with high quality.

Detecting with accuracy an object by a robot is linked to images captured, light applied to the object, its position in space, occlusion, its movement if it is necessary to perform recognition dynamically and other factors that are dependent on each system. A performance object tracking software must meet a list of requirements to be somehow close to the human visual system. Features of software found in this lineup includes tracking multiple objects, image stabilization, fast detection speed, flexible architecture which is required for a robot which is in continuous development. In this article we gathered together six of the most used object tracking software in robotics.

1. SwisTrack

SwisTrack Multi-Object Tracking Software

SwisTrack is one of the most advanced software used for multi-object tracking in robotics. It can be used for tracking objects, other robots, animals, humans, etc. As input it is required to use a camera or a video recorder. Using OpenCV library and having a flexible architecture, SwisTrack can be used for tracking objects in different situations. It has support for USB, FireWire, and GigE cameras. Continue reading (…)

A Step Closer to Humans – Artificial Skin for Robots

Robots have been around long enough for a rich research environment to be created where Sci-Fi movies have become reality. The human body is one of the main instruments of inspiration for researchers, and the human senses have become a means to enrich the robotics technologies. The human skin has many roles, but only one of them is implemented into intelligent machines using materials such as silicon or carbon.

An intelligent robot is a robot that feel and interpret the environment. Without sensors this metal box could not have a well established role. A cup, an egg, or a pen have different forms and can withstand without damage different gripping strengths, thanks to their outer shells. For a robot is impossible to feel and manipulate these objects without a wide range of sensors including reading of force, pressure or temperature. In this article you can find a list of sensors, an artificial skin consists of a large number of flexible and thin sensors, that can be used to dress up a robot from head to foot.

01. Seoul National University Electronic Skin

Seoul National University Electronic Skin (Photo: Gizmag)

Researchers at the Biomimetic Laboratory of Seoul National University have imitated the nature and built an artificial skin with a simple design and a high accuracy. The electronic skin is built from two layers of polyurethane acrylate with a low-cost production process. The inspiration comes from a system that can be found in an ear and can interpret the sounds to transmit signals to the brain, the cilia. Continue reading (…)

Industrial Robots versus Service Robots

Like the Olympics where athletes from around the world compete in various sports categories, Robotics is a field where researchers from all continents compete in the creation of innovative technologies. The progress of intelligent machines and their adoption in many fields of activity led to a clear classification of robots. Furthermore, this entire classification distinguishes between two general categories: industrial and service robots. Each of these two general types of robots are divided into categories based on the activity fields, functionality and market demands. Both, industrial robots and service robots, can be divided into two categories each. In industry stationary robots and mobile robots are employed, while service robots can be split into personal and professional service robots. Even though the first industrial robot and the first service robot were created in different periods of time, service robots have had a much greater impact on the robotic market entirely, and have an important share of it.

The first robot, called Unimate, was built in 1961 and was used by General Motors. This robot was used in industry and is the father of the robotic industry which is growing rapidly nowadays. Industrial robots are highly specialized and are destined for various areas of manufacturing goods, from industrial robots used to build cars to robots used to manipulate materials in warehouses. These are automatically controlled machines, can be reprogrammed, and exist in different variants and designs which can be classified according to the number of axes – three, four, five, six, or more. The biggest problem of industrial robots is actually a general problem in the production of goods – the logistics. This problem found a solution in autonomous industrial mobile manipulators. Continue reading (…)

An Overview of Affordable Robotic Arm Kits – Part 2

In the second part of our review we will talk about robotic arm kits that feature more slightly more advanced construction and control characteristics, making them equally well suited for research purposes as well as for more complex hobby applications. These kits retain affordability while providing versatility and robustness, allowing for easy customization to integrate with your applications and precise control leading to improved performance. In part 1 of our review we also briefly presented key aspects of robotic arm physical constructions, this time we will start by addressing, in a succinct manner, principles of numerical representation and control of robotic arms.

Forward kinematics (FK) and inverse kinematics (IK) are two methods for positioning of a robotic arm and its end effector, i.e. gripper.

  • In FK final position and orientation of a gripper can be established based on known joint angles and length of links or sections of a robotic arm. This is a pretty straightforward method, relying on trigonometry and basic mathematic calculations. In other words we can estabish where the gripper is when the robotic arm is in a certain position determined by its joint angles;
  • Conversely, IK is used to establish positions of joints in a robotic arm when the position of the gripper is known. Since link lengths between joints are known, the method is used to calculate joint angles required to reach that desired position of the gripper. This is a much more complex calculation method which can absorb an important amount of resources, especially in microcontrollers where processing power is at a premium. The method is especially difficult since zero, multiple or even infinite configurations of a robotic arm may exist for a single certain position of the end effector, involving multiple exponential equations. In this case a positioning algorithm must take into account many factors such as previous positioning data, physical characteristics of the arm and so forth. Thankfully several tools exist for developing fairly capable IK algorithms suited for microcontrollers employed by the various platforms presented in our roundup.
PhantomX Pincher robot arm

Photo: Trossen Robotics

Some kind of motion planning must also be performed in order to optimize motions of a robot. A simple method would be to consider a desired path or trajectory of a robot and decompose it into simple geometrical forms such as straight lines, circular arcs or angles. Velocities and joint forces must also be considered by taking into account weights of each element of a robot, actuator characteristics and path travelled by gripper to determine response time and optimum behavior. Robotic arms need to be as rigid as possible while maintaining lightweight in order to reduce or cancel parasitic movements, characteristics exhibited by the platforms presented below. Continue reading (…)

Real Life Robot Games set a New Revolution for the Gaming Industry

Just like work and relationships, games are an integral part in the social life of every human and have a special contribution to the development of imagination and direct physical activity. Computer games took a strong momentum in the last decade in the detriment of offline games. New technologies like Kinect from Microsoft or Wii from Nintendo opened again the way to offline games, interactive games involving movement. Still in an early phase, games that involve robots have a great potential for the future wave of gaming industry. These games involve more than one screen, are played between humans and intelligent machines, you can say its a game between friends.

Regardless of age, robots could be our next playmates. Companies producing robots began to feel the need of such technologies which involve both physical activity and imagination. Unlike games like Wii, a robot can be taken to the park, on vacation or wherever you wish to interact with it. The mobility of these smart devices brings value to games and creates a space large enough for future games which will certainly be full of intelligence.

Playing with a friend is full of emotions, a missing detail for a robot. For example industrial robots are full of muscle and follow strict rules of operation. Social robots have a different purpose, they are designed to interact and listen to people’s desires. The perception of such intelligent devices may be different from region to region, depending on culture or religion. In Japan, the most robotized country in the world, I would say that a robot is an additional member of a family. In general, Asian countries are closer to those smarter pieces of plastic and metal than the rest of the world. The facts are different for the the old European continent as well as North America, where people are a little reluctant to communicating with such robots.

Robot interaction methods are important factors during the game. Combining mobile devices with robots led to a huge success among robotics enthusiasts which develop applications for robots control. Employing either a smartphone or a tablet, the robots can be controlled and can interact with the user during the game. Luckily such applications exist for both major mobile OS’s so users are not bound to neither to Android phones or tablets nor Apple’s iOS devices. For starters, as such devices and applications are still emerging, there are a limited number of robots which can be used to play in the real world and less in the virtual world.

Parrot AR.Drone

Parrot AR.Drone

The Parrot AR.Drone is a quadrotor flying robot which can capture and transmit in real time photo and video images on a smartphone or a tablet remotely. The images are of the highest quality with 720p HD resolution. To be connected with robot the first step is to install an application called AR.FreeFlight 2.0. The flying robot can be used both indoors and outdoors. Any user can control the aircraft in order to capture images or for entertainment. Manufacturing company offers a variety of games that involve a direct use of drone. Continue reading (…)

An Overview of Affordable Robotic Arm Kits – Part 1

Today robotic arms are employed on a large scale, for an endless list of applications. You can find robotic manipulators almost everywhere, in industrial environments performing highly repetitive or heavy duty manufacturing activities on assembly lines, in research facilities precisely handling sensible or hazardous substances or objects and even on other planets gathering samples or otherwise. There are few activities a robotic arm cannot handle, successfully increasing productivity and precision or seamlessly operating in environments not suited for humans. Robotic arms are generally designed to replicate a human arm so they are relatively similar in structure, however there are countless variations of design depending on applications for which they are specifically built. Compared to a human arm robotic arms can have reduced or elevated complexity and abilities in terms of movements they can perform and loads they can carry.

Dagu robotic arm

Photo: Dagu Electronic

A primary criteria that defines a robotic arm is the number of degrees of freedom (DOF), each degree of freedom representing a motion a robot can perform. For a robotic arm to perform a motion it requires a joint which allows either rotational or translational movement, and of course an actuator to power that section, therefore the number of DOF can be easily determined by establishing the number of actuators in a robotic arm. The more DOF the better, as the robotic arm can perform a greater range of movements. Each DOF however has a great impact on the complexity, not necessarily in terms of the robotic arm’s structure but rather in terms of the mathematical models describing it. These models can be extremely complex and can absorb a huge amount of resources so it is recommended to implement the minimum amount of DOF possible required for an application as everything, needs to be taken into account when developing motion algorithms in order to reduce unwanted or potentially hazardous operation to a minimum.

Another criteria is the workspace of the robotic arm which is comprised by absolutely all positions its business end, i.e. gripper or tool that directly performs a task, can occupy or reach in an environment. This is important as the robot must be able to reach what it should operate onto, as well as stand clear of everything else in the environment. Other factors defining a robotic arm are extremely dependent to application and include sensor arrays, load capacity, joint forces, power sources or types of actuators. For more information you can read our dedicated articles about industrial robots.

Although robotic arms employed in industrial or otherwise professional environments can be in the price range of tens and even hundreds of thousands of dollars, slightly out of reach even for some institutions let alone a home user or developer, the market for light duty affordable robotic arms is continuously growing as robot manufacturers have developed budget alternatives for hobbyists or professionals alike. In this series of articles we will discuss about sub-1000 dollar robotic arm kits. Continue reading (…)

New and Innovative Technologies used for Robots

Researchers are not on vacation and this can be demonstrated by innovative technologies which can be used in robotics. The robotic field is complex and takes time to create new technologies that lead to a solid growth, technologies inspired by nature where solutions had millions of years to reach a high degree of perfection. Sci-Fi media has an important role in breaking conceptual barriers that may occur as a form of capping a project, and in time fiction can become true technology.

An analysis of new technologies which can be used in robotics reveals the trend which will be followed by the next generation robots. These technologies include artificial muscles, elastic electric cables, artificial skin, artificial brain, fast moving robots, artificial noses or tactile finger devices and have the potential to bring artificial devices one step closer to living organisms and even humans.

01. Artificial muscles and actuators from carbon nanotubes

Carbon is a chemical element, symbolized by the C letter, and is one of the most abundant element in the Universe. Carbon nanotubes (CNTs) are allotropes of carbon with a cylindrical nanostructure. Through the power of creativity, researchers use carbon nanotubes to create artificial muscles and actuators. It is not easy to create muscles that are able to contract and twist from such materials, especially when the inspiration from nature is limited in this case.

Artificial muscles and actuators

In this project the researchers have created artificial muscles by using a single-walled carbon nanotube (SWNT). SWNT is a complex structure which includes double electrode layers separated by a chitosan electrolyte layer consisting of an ionic liquid. Thanks to special properties of the SWNT structure, actuators made from this material have orders-of-magnitude improved characteristics compared to previous ionic electroactive polymer (i-EAP) actuators. Continue reading (…)