Are you looking for a comprehensive guide on robotics for beginners, and want to know about everything which goes behind the development of a sophisticated robot?
Or, do you just want to learn about the basics of robots and develop your robot from scratch?
Maybe, you are just looking for an extended answer to a very simple question, “What is Robotics?”
Well, whatever the reason may be, I assure you that you have made the right decision to learn about the robots and that too, at the right time…
COVID-19 has exposed the vulnerability of humans to various unknown diseases. Besides, extended lockdown periods have resulted in the closure of many manufacturing facilities.
As a result, many industries have discontinued their sole reliance on human workers, and allocated hefty budgets to automate critical sections of the manufacturing facilities.
From a business perspective, 88% of the businesses worldwide plan to adopt robotic automation into their infrastructure.
These stats indicate that Robotics is the Future, and you are on the right learning track!
The real question is, “Where should you start?”
Let me tell you my personal experience first…
When I first started to explore the basics of robots, I didn’t know that field of robotics comprised of Electrical Engineering, Mechanical Engineering, Mechatronics, and Computer Science.
Moreover, understanding blended information from all these fields was never going to be an easy task for me. Thus, I had to go through a large number of blogs and research articles to understand the different components which make up a robot.
As a result, it took me a lot of time and effort to get the slightest understanding of how robots work!
You may be thinking that you also need to do the same…
I have compiled everything in this post, which you will ever need to get started with the robots…
Let’s start with the most basic of the definitions
What is Robotics
Robotics is an interdisciplinary branch of science, electrical engineering, mechanical engineering, and computer science that deals with the physical motion of the robots.
In simple words, it involves everything from conception to mechanical design, manufacturing, and control of robots to carry out a sequential set of tasks autonomously or semi-autonomously.
As simplistic as the definition may seem, there is still some confusion regarding the term ‘Robotics’ among the general audience. The main reason being that people tend to confuse the term ‘Robot’ with ‘Robotics’.
To be precise, the knowledge of ‘Robotics’ is primarily used to MOVE a ‘Robot’ within the physical space of the humans, which involves designing, building, and programming the robots.
On the contrary, a ‘Robot’ may also have many other sub-knowledge components i.e., Artificial Intelligence for reasoning/inference, Machine Learning for pattern recognition, and Natural Language Processing for real-time communication, etc.
By now, you may be wondering, “What is the difference between Artificial Intelligence and Robotics, then?”
Believe it or not! This question alone creates more confusion among the novices than the sophisticated robot itself.
Anyways. Let me clarify it once and for all…
Artificial Intelligence in Robots
There is no doubt that most modern robots are equipped with artificial intelligence, which empowers them to make inferences and draw conclusions.
However, if a robot is deprived of such intelligence, will it still be a ‘Robot’?
The answer is a Simple YES!
A robot is built to assist humans by automating day-to-day tasks. And, a robot can surely do that without requiring even a pinch of intelligence!
Let me explain this concept using a simple example of a pick-and-place robot. In the majority of the applications, this robot will pick up a component from one assembly line, and place it onto another assembly line.
This simple repetitive motion can be fed into the robot’s controller, and the robot can do that job all day long!
More importantly, there is no intelligence being shown here by the robot. It just moves between the two points, which are also dictated by a human operator…
Thus, the knowledge of ‘Robotics’ is in action here, but not that of ‘Artificial Intelligence’!
Now, let’s consider another use case of the same robot…
If a factory is manufacturing footballs of three colors, and there are different packing boxes for each color of the football, then a simple pick-and-place movement will not suffice.
In this case, the robot firstly needs to pick the football, inspect its color, and then place it in its respective packing box. Hence, the use of a pattern recognition algorithm is imminent here, and the robot will be called an Artificially Intelligent Robot.
I hope this simple example clears any confusion between Robotics and Artificial Intelligence. Before moving forward, I have a simple assignment for you!
Can you think of any other scenario which is well suited to explain the perceived difference between Artificial Intelligence and Robotics? If yes, then mention it in the comments section…
Robots vs Machines
Before I formally start the comparison, I need you to ponder over this very simple question:
“Why is a sewing/washing machine called a MACHINE, and not a ROBOT?”
From a functional perspective, both the machines and the robots have only one basic purpose: To Assist Humans in a designated task…
However, the primary difference between them lies in the underlying Control Framework.
The Control Framework simply depicts how the Current Output of a system is aligned with its Desired Output.
Mainly, there are two types of Control Frameworks:
Open-Loop Control System:
If a human is needed to align the current output of a system with the desired output, then such a system is called an Open-Loop Control System.
Manual control of Fan Speed using the dimmer is an example of such a system. A human is needed to regulate the current fan speed as per the desired value.
Closed-Loop Control System:
If a system can sense the current output of a system while enforcing itself to achieve the desired output without the intervention of any human, then such a system is termed as a Closed-Loop Control System.
Automatic Fan Speed Control is a closed-loop control system. A human can set the desired value (e.g., 3500 RPM), and the fan speed will decrease/increase automatically to achieve the desired speed.
In the context of Robots vs Machines, a Robot is a Closed-Loop Control System, whereas a Machine is an Open-Loop Control System.
Let me clarify this concept using more examples from daily life…
When you need to blend some ingredients, you put the ingredients in a blender and let it run for 1-2 minutes.
After that, you check the mixture and if the ingredients are not blended properly, you run the Blender again. You repeat this process until you have got the perfect mixture.
In all of this process, you checked the output of the Blender yourself. There was no system installed in Blender to adjust the mixture automatically as per your liking. Hence, the Blender is a MACHINE.
In an alternative scenario, what will happen if you turn on the Automatic Vacuum Cleaner?
The cleaner will sense the dirt itself, and then suck it off your floor; you don’t need to guide it regarding the whereabouts of the dirt. Moreover, it will traverse the whole room without any specific guidance from your side.
In simple words, a cleaner will do its intended tasks without any assistance from a human. Hence, the Automatic Vacuum Cleaner is termed as a ROBOT.
Three Laws of Robotics
Due to the movement of robots within the physical space of humans, there exists an imminent threat from robots to harm fellow human beings.
To mitigate this threat, Asimov suggested the Three Laws of Robotics which guide the behavior of robots within a human-cluttered environment:
- Robots must never harm human beings
- Robots must follow instructions from humans without violating the first condition
- Robots must protect themselves without violating the first two rules
While developing the movement patterns of any robot, or during its installation in any manufacturing facility, it must be ensured that the robot complies with the Three Laws of Robotics.
This practice is necessary for seamless integration of the robots within the human surroundings to produce an overall effective and collaborative environment.
How Robots are Made
When you see a sophisticated robot in action, a simple but intriguing question may pop up in your mind: “How the hell this Robot was made?”
To your surprise, this question has the simplest answer…
There are over 7.5 billion people living on Earth. If you look at the human body, it mainly consists of a skeleton, internal organs, muscular tissue, and body fat. However, the intelligence level of all human beings differs from each other.
For instance, an intelligent being may solve a specific problem within seconds, while a normal human may not be able to solve the same problem at all.
A similar approach can be used to align the operational complexity of different robots with an underlying structural framework.
In simple words – As complicated as the operation of a robot may seem, the development of any such robot primarily consists of 3 main modules:
1. Mechanical Construction
Just as a human possesses a skeleton to move around, a robot also needs to have a necessary mechanical structure to interact with the environment and get the work done.
However, unlike humans, the mechanical structure of each robot differs as per its intended use. In simple terms, a robot can only perform a specific task for which it is designed.
For example, a MARS Rover has a different structural framework than a Pick-and-Place industrial robot. A Rover can move around in an unknown environment while inspecting different elements in the surrounding atmosphere. On the contrary, an industrial robot can perform the pick-and-place operation in a predictable environment only.
Moreover, the material used for the manufacturing of a robot also differs because of intended loading requirements as well as the overall weight of the robot.
For example, a weight-lifting robot manufactured using PLA (via a 3D printer) may lift the loads to 1kg. However, if an application requires the same robot to lift weights up to 30 kg, then different materials will be used to satisfy the load requirements i.e., Iron, Steel 7075, Carbon Fiber, etc.
For the application of wearable robotics (prosthesis and exoskeletons), it is desired to minimize the robot’s weight and maximize the structural strength.
In such cases, a material with high tensile strength and low density is the recommended choice. For example, Carbon Fiber will be preferred over Aluminum, and Aluminum will be preferred over Steel.
Before diving into the manufacturing process, you need to design the whole mechanical assembly in a 3D modeling-based CAD software i.e., SolidWorks, AutoCAD Mechanical, Solid Edge, etc.
Once the mechanical design is finalized, you need to simulate a Stress Analysis test on each part of the assembly. This will help you analyze whether the mechanical assembly can bear the stresses imposed during the operation or not.
If the Stress Analysis tests turn out to be successful, you need to prepare the assembly drawings, and then pass them over to a fabricator to start the manufacturing process.
After the mechanical assembly is manufactured and fitted, you need to make arrangements to power and move the assembly using electrical components…
2. Electrical Components
A human with a bare skeletal structure cannot perform any action, even if it’s a simple finger movement. Muscles provide the necessary energy to the human skeleton, whereas Internal Organs are responsible to produce that energy from the ingested food.
Similarly, the mechanical structure of a robot cannot do anything on its own. It needs an external power to move and complete its intended task.
You can also move the mechanical structure all by yourself, but that’s not the reason you built this structure in the first place, right?
So, what’s the solution?
You need to integrate Electrical Components within the mechanical assembly to bring it to life. These components power the structure as well as provide a ready means to control it.
From an application perspective, there exist 3 broad categories of electrical components:
- Power Source
Power Source for a robot is just like food for human beings, or fuel for cars. It provides the necessary power reservoir, which the actuators and sensors can utilize to manipulate a robot.
Actuator takes input from the Power Source and converts it into the physical movement of the mechanical structure e.g., wheels for movement, gripper movement for pick-and-place, etc.
Robotic Sensors mimic the Five basic human senses. They allow the robots to recognize and inspect their surroundings e.g., atmospheric temperature, soil humidity, size/shape of the object, obstacles along the path, environment visualization, human speech recognition, etc.
Once the mechanical structure is equipped with the electrical components, you have got yourself a robot that can move around in an environment as per the commanded instructions.
However, the real question remains to be answered…
“How to Provide Instructions to a Robot?”
It’s simple. You need to provide Digital Instructions to the robot, and Computer Programming is the best way to do that…
3. Computer Programming
If you thought about lifting your arm, but your body responded by lifting your leg instead, then who is at fault here?
Your MIND, indeed…
Your skeleton, muscles, and internal organs have done their job alright. But it was your MIND which messed up the things.
In plain terminology, your MIND controls how your body behaves in any particular situation. It takes the inputs from your basic senses and instructs your body to produce the desired response.
In similar terms, the mechanical structure and electrical components only equip the robot to perform different tasks and record environment variables.
You still need to tell the robot about Which Actions to Perform and How to Perform thembased on the sensory data…
However, here is another twist… You cannot just shout at the robot, and ask it to complete a task.
The robot won’t even understand your language. You need to communicate with the robot in a language it can understand – The Binary Language!
Normally, a binary program is embedded within a microcontroller to establish digital communication with the robot.
Binary data is collected from the sensors installed on the robot and then processed subsequently. In the final phase, actuators are fed with the appropriate binary commands to produce the desired response.
The program code is usually written in high-level languages such as C/C++, Python, MATLAB, etc. A compiler then converts the high-level program code into a binary (machine) code, which is fed into the microcontroller.
Most frequently used Microcontrollers in robotic projects include Arduino, PIC, STM32F4, Raspberry-Pi, Beagle bone, and Jetson Nano, etc.
The microcontroller must be chosen as per the hardware specifications and the intended use of the robot. For example, if you just want to control an obstacle avoidance robot with 2 wheels, 3 IR sensors, and 2 UV sensors, an Arduino would be a far better choice than the Raspberry-Pi.
Now that you have learned about the structural and control framework of different robots in general, it’s time for you to deepen your understanding of the components which power most of the sophisticated robots in this world.
Let’s move on…
Main Components of Robots
You may have heard of an aphorism,” All Five Fingers are not the same”.
Well, the same rule applies to the world of robotics, and it states that All Robots are not Created Equal.
The major components used in robotics have different variants because of diversified applications.
To get started with robotics, you must be aware of the majority of them…
The power reservoir of any robot energizes the electrical components within a robot, and it classifies as the foremost need for the operation of any robot.
Different variants of power sources exist to cater to different robotic applications:
1. Power Supply
This power source takes AC Power as input and produces DC Power on the output. It is the most commonly used power source during the development and testing phases of the majority of the robots.
One of the main advantages of the Power Supply is the stable DC Power output, with minimum-to-no ripples in current or voltage. This ensures safe operation for actuators and sensors sensitive to voltage variations.
In sophisticated power supplies, you can also control the current being drawn by the electrical components. This feature proves beneficial in operating the actuator within a continuous current range to avoid any permanent damage.
An inherent disadvantage of the power supply is the need for a continuous supply of AC input. This limitation forces its use for the robots which are fixated at one place i.e., Industrial Robots.
However, it can be also used for mobile robots, albeit for testing and development purposes only. In such a case, a power connection must be tethered to the mobile robot all the time which may also restrict some of its movements.
A battery is a portable alternative to a power supply. In simple words, a power supply needs a continuous supply of AC Input, whereas a battery does not impose any such restrictions.
You need to charge a battery to its full capacity, and then use it to power the robot until it gets discharged. In that case, you will need to charge the battery again and this cycle goes on.
Because of the portability, the battery is a preferred power source for mobile robotics. However, the charging/discharging cycle of a battery disrupts the workflow during the testing or development phase.
Hence, it is a recommended practice to use a tethered power supply during the testing/development phase, and a battery during the deployment phase.
While selecting a suitable battery for your robot, you must always strive for a high power density. The greater the power density, the greater will be the amount of power delivered per unit mass of the battery.
A mobile robot needs to carry along a battery pack, and its additional weight may limit the operational capability of the robot. Hence, you must ensure that maximum power is delivered to the robot for as low weight as possible.
3. Solar Panel
Solar Panel is a renewable alternative to a power supply and/or a battery. It takes Solar Rays as input and produces DC power on the output. The larger the size of the Solar Panel, the greater the power wattages produced.
A major drawback, in this case, is that output power can only be produced under solar light. Hence, a robot primarily drawing its power through a solar panel may not be a good choice for indoor operations or during the nights.
But, Wait! There exists a turnaround for this drawback…
While the robot is operating under the sunlight, you can divert the power drawn from the solar panel into two pathways; one pathway will operate the robot, whereas the other pathway will utilize the excess power to charge the battery installed within the robot.
Once the sun goes down, you can draw the power from the battery to operate the robot. In this case, however, the robot will need to carry the extra weight of a battery, plus the battery charging will only last for a definite period.
Moreover, the power drawn from the solar panel is portable and does not need a tethered power connection.
More or less, solar panels are suitable for low-power applications because the robot will only need to carry a little extra weight of the solar plate.
4. Internal Combustion Engine
You have already seen an engine as the real driving force behind the working of automobiles.
In essence, an Engine is just a power source that takes Fuel as input and produces Mechanical Motion as the output.
In automobiles, mechanical motion is used to rotate the wheels, whereas the same motion can be used to operate the robots as well.
An Internal Combustion Engine is a preferred choice for high-end applications, where an extraneous amount of mechanical power is desired. This is why you don’t normally see an engine driving a robot having optimal power requirements.
As the world is already migrating towards the use of renewable energy sources, there is a slight chance that you will get to see a robot using an engine as its primary power source.
Pneumatics make use of an air compressor to reduce the volume of gas in order to increase its pressure. This high-pressured gas then applies power on a moveable piston, which in turn produces a mechanical motion.
The force exerted by the mechanical motion depends on the volume of gas being compressed, as well as the compression level itself.
Since air can be easily compressed, pneumatic systems can easily absorb excessive shocks and are mostly used to actuate smaller loads.
Hydraulics uses the same underlying technology as pneumatics but makes use of liquids rather than gases.
The operation of hydraulics is primarily governed by Pascal’s Law. As per this law, when a liquid is pushed from a wider area to a narrow one, the speed of the liquid increases at the expense of less force exerted on the narrow end.
On the contrary, if a liquid is pushed from a narrow portion to a wider one, the speed of the liquid decreases but it exerts a magnified force on the wider end than originally applied at the narrow end.
Since a magnified force is applied at the output (wider end), hydraulics is mostly used to actuate heavy loads where speed/acceleration is not much of an issue.
The whole concept of actuators is to take input from the Power Source and convert it into a mechanical motion.
The output of the actuator can be a linear or a rotatory motion. However, both of these motions can be interchanged just by adding some extra mechanical assembly.
Actuators differ by the type of input they can receive from the power source, as well as by the type of mechanical motion they can produce on the output:
1. Electric Motor
An electric motor converts Electrical Energy into Mechanical Energy. In simple words, it takes electrical power as input and produces mechanical motion as the output.
Using this convention, an electric motor can receive input from three power sources namely Power Supply, Battery, and Solar Panel.
At the output, it produces a rotatory mechanical motion on the shaft. The shaft is further attached with the Wheels or Links to induce motion into the robotic system.
Output Torque and Speed of the electric motor is directly proportional to the Current and Voltage being supplied by the power source, respectively.
Electric Motor is the most frequent actuator being used in robotics because of a flexible selection choice from its different types i.e., PMDC Motor, BLDC Motor, AC Motor, Stepper Motor, Servo Motor, Geared Motor, Industrial Grade Motor, etc.
Apart from that, you have the flexibility to choose from different Speed-Torque Characteristic variants of a single type of motor.
For instance, you can select a High-Torque/Low-Speed BLDC motor to lift a heavy load, whereas a High-Speed/Low-Torque motor for an electric fan.
Give or take, you just cannot ignore electric motors in the field of robotics. If you are a beginner in robotics, you don’t even need to look beyond the motors in your starting phases.
2. Pneumatic Actuator
A pneumatic control valve actuator converts the force exerted by the high-pressure gas into a mechanical motion.
This actuator can only use Pneumatics as the power source. It normally consists of a piston followed by a hollow cylinder.
Due to increasing or decreasing gaseous pressure, the cylinder moves along the axis of a piston thereby creating a linear mechanical force. This force is then used to produce motion in the robotic assembly.
3. Hydraulic Actuator
A hydraulic actuator operates in a fashion similar to that of a pneumatic actuator, but an incompressible liquid is forced from a narrow to a wider portion of the pump to move a piston attached with the cylinder.
The hydraulic actuators are suitable for high-force applications and only use Hydraulics as their power source.
According to one estimate, hydraulic actuators can exert a force equivalent to 25 times than a pneumatic actuator of the same size.
4. Linear/Rotatory Actuator
Linear or Rotatory actuator is a classification of the actuators based on the type of mechanical motion produced at the output i.e., linear or rotatory.
A linear motion is a simple forward or backward movement, whereas a rotatory motion is a complete 360 Deg. movement along the axis of the rotating shaft.
Each of the electric, pneumatic, or hydraulic actuator can produce either linear or rotatory motion. The selection of a suitable actuator is specific to the application at hand.
Sensors allow the robot to take in the information from the surrounding environment, as well as the internal components of the robot itself.
In simple terms, sensors are the EYES and EARS of a robot. They guide the controlled behavior of robots among so many distractions in a human cluttered world.
With the help of a sensor, robots can easily perceive changes in an environment and act accordingly.
Sensors mostly differ by the type of information being collected from the environment:
1. Motion Sensors
These sensors give information to the robot about its movement. This information can be applied to adjust the current movement patterns of the robot against the desired values.
Since a robot firstly needs to know its position before commencing any further plan-of-action, you are more than likely to see these sensors in almost every robot.
To detect the position and speed of the robot, you need to convert the mechanical displacements of the robotic manipulators into electrical signals. And fortunately, that’s what Encoders do!
A value of Ticks Per Revolution is specified for each encoder; this is known as the Encoder Resolution. You can calculate the number of Ticks during the robot’s manipulation, and divide it by Ticks Per Revolution to get the total number of revolutions (1 Revolution = 360 Deg).
For example, if the Ticks Per Revolution is specified as 2048 and you have calculated 1024 Ticks during the operation, then the robot has only moved 180 Deg.
Different types of encoders exist to cater to the sophistication required in specific applications i.e., linear, rotatory, incremental, absolute, optical, etc.
Inertial Measurement Unit (IMU)
IMU consists of a gyroscope, accelerometer,and magnetometer. When installed within a robot, it can measure a variety of factors including speed, acceleration, direction, angular rate, inclination, and orientation of the robot.
In the case of the magnetometer, IMU can also measure the magnetic fields surrounding the robot. It helps in the GPS positioning systems of the robot.
Each component in IMU is responsible for the measurement of different parameters:
- Accelerometer: Measures Velocity and Acceleration
- Gyroscope: Measures Rotational Degrees as well as the Rotational Rate
- Magnetometer: Measures magnetic fields and establishes a directional heading
The output from all the components can be combined to formulate an accurate movement of the robot in 3D space.
2. Vision Sensors
These sensors help robots visualize and inspect their surroundings during operation. With the help of vision sensors, robots can detect objects and traverse safely while avoiding obstacles along the way.
These sensors are only necessary if the robot needs to make sense of its surroundings. For example, a mobile robot traveling through a neighborhood must have vision sensors, whereas an Industrial Pick-and-Place robot does not need any vision.
The camera is the most-ready source of inculcating vision into a robot. The robot receives a frame-by-frame video stream of its surroundings through a camera lens.
Each frame can be processed to extract useful information i.e., Object Recognition, Obstacle Position, Face Recognition, Gesture Recognition, and Color Detection, etc.
The output interface of the camera poses problems in its integration with different robots (or microcontrollers). For example, a robot using Raspberry-Pi may integrate well with a USB Camera, but a robot controlled by Arduino will need a different camera module.
An infrared (IR) sensor is used to detect objects in the surroundings of a robot. IR sensor works in the Infrared Spectrum rather than the visible spectrum of light.
In the Infrared Spectrum, each object emits some form of thermal radiation. A human is unable to see this radiation, but an IR sensor can surely detect it.
An IR sensor normally consists of an IR Transmitter and an IR Receiver. IR Transmitter is simply an LED that emits infrared light, whereas IR Receiver is a Photodiode which detects the light in the infrared spectrum only.
As the robot moves, the IR Sensor keeps on emitting and receiving the infrared light. If an object is close to the robot, the intensity of the received IR light does not match with that of the emitted one.
Rather than installing just a single IR sensor at the front of the robot, it is preferred to install them all around the robot. In this way, a robot can detect and avoid all the obstacles surrounding it.
3. Environment Mapping Sensors
While vision sensors are used to get feedback about the nearby surroundings, environment mapping sensors are used to map and localize distant objects in an environment.
These sensors are only used for high-end applications, where the robot needs to operate in a highly unreliable and unstructured environment.
Unlike vision sensors, these sensors can detect the three-dimensional position of different objects in the visible environment. Not only that, they can localize the object as well i.e., specify its relative position w.r.t the robot.
Light Detection and Ranging (LiDAR) is a remote sensing technology to measure distances (relative position) of different objects in an environment.
It illuminates the environment with laser light and measures the reflected light with a laser scanner.
The difference between the transmitted/reflected laser light in terms of Return Time and Wavelength can be used to formulate the 3D representations of the environmental objects.
Radio Detection and Ranging (RADAR) system is used to detect, track, and locate objects at considerable distances.
A RADAR emits radio waves in a specific direction and then compares the returned echoes with the emitted ones to make sense of the objects in the environment.
For example, the time difference between the emitted and reflected echo can be used to measure the distance (relative position) of the object.
Similarly, the timestamps from the two-consecutive transmitted/reflected echoes can be used to calculate the velocity of the traveling object.
RADAR can also determine the range (distances) of distant objects with precision, even in adverse weather and environmental conditions.
Sound Navigation and Ranging (SONAR) is used to detect underwater objects, as well as their distance and direction.
An acoustic projector generates a sound wave that travels away from the source. If an object is present in the path of the sound wave, a part of it is reflected.
A receiver picks up the reflected wave, analyzes it, and extracts useful information about the detected object i.e., range, direction, distance (relative position), etc.
SONAR is mostly used for the navigation of underwater robots, or the robots responsible for the detection of underwater objects.
4. Application-Specific Sensors
Each application requires a specific set of sensors, and I can’t explain each one of them.
In a broader perspective, each sensor does the same thing; sense something, and then represent it in terms of electrical signals. These signals can be communicated to the robot’s controller via analog or digital communication protocols i.e., I2C, SPI, UART, USB, etc.
Here are some more sensors which you may encounter in common applications:
- Humidity Sensor: To measure the Moisture Level in the air
- Temperature Sensor: To measure the temperature of the environment
- Liquid Pressure Sensing: To measure the Pressure of liquids or gases
- Light Sensor: To measure the Intensity of light
- Force/Torque Sensor: For the monitoring of Linear or Rotational Forces on the robot’s End-Effector by the environment
- Current Sensor: To measure the Current (Mechanical Torque) of the actuator
- Mic: To get the Voice Input from a Human
If you get your hands dirty with the commonly used sensors, you won’t have any problems interfacing with the other types of sensors.
If you ask me about one topic in the field of robotics, which is not clearly defined and has no set boundaries altogether, then my answer will be the Types of Robots.
Just imagine all the tasks a human can perform, and then try to estimate the types of robots needed to replace each human task.
How much did you get?
Well, this is where the confusion prevails. Every article on the internet tries to set its boundaries for the types of robots, but still misses a few!
Now, I am also going to do the same. I hope to cover each robotic type, however, if I miss something, you are more than welcome to give your input…
In a broad sense, robots can be categorized based on two primary characteristics:
Now let’s define sub-types under these two main categories…
1. Types of Robots based on FUNCTIONALITY
Functionality differentiates the robots based on the unique characteristics a robot possesses as well as specific functions a robot can perform.
In this category, each robot type must possess one unique characteristic out of all the other robotic types…
These robots are programmed in-advance by a human operator to perform a monotonous task repetitively.
Such robots are mostly used in an Industrial Setting for the tasks, which otherwise would require an extraneous effort from a human worker i.e., painting, welding, pick-and-place, etc.
An Industrial Robot has the added advantage of working continuously 24/7 in comparison to a human worker, who is most likely to get exhausted after an 8-hour duty.
The base of an Industrial robot remains fixed at one place; however, it operates by extending the interconnected links attached to the base. In most cases, the work area of the robot is defined and human workers are strictly prohibited from entering that area while the robot is operating.
The next-generation of industrial robots are known as Collaborative Robots because they can work in collaboration with a human worker to maximize the productivity of the assembly line.
The robots which use LAND as their operational base fall under the umbrella of walking robots.
Walking robots can be further categorized according to their movement:
Wheeled Robots are the ones that use two, three, or four wheels to move forward/backward as well as to change their direction e.g., robotic vacuum cleaner
Quadrupled Robots have four legs, and use a gait mechanism similar to four-feet animals. Balancing is usually not an issue in quadrupled robots, but the synchronized movement of the 4 legs poses problems e.g., MIT mini cheetah
Bi-pedal Robots have two legs just like humans. They are also known as Humanoid Robots because they are human look-alike and exhibit human behavior. They can perform human-like activities i.e., running, jumping, etc. Since a Humanoid Robot only has two legs, balancing of such a robot during its operation classifies as an inherent control problem. The famous Humanoid Robots include Honda’s Asimov, Boston Dynamics’ Atlas, and SoftBank Robotics Nao.
Flying Robots use the medium of AIR to fly around the atmosphere to perform their tasks e.g., Drones, Quadcopters, etc.
The tasks may include sweeping over a certain area for security reasons, or flying around the agricultural land to detect and report back any anomaly in the crops.
In most cases, a flying robot is equipped with a camera which allows for the recording of the under-consideration environment from a distant field of view.
The resulting footage is then subjected to Image Processing techniques to extract useful information i.e., dry patches in a crop, a group of people violating the COVID-19 SOPs, any unusual activity from the security point of view, etc.
On a lighter note, you can also use flying robots to record your FUN Activities or just for Aerial Filming…
These robots either operate on the surface of the water (sailing) or under it (swimming). For the robust operation of such robots, electronics contained in them must be water-proofed via an air-tight assembly packaging.
Major applications of such robots include maintenance, inspection, and surveillance of underwater pipelines and structures.
These robots use SONAR to navigate successfully under the ocean and to communicate critical messages with a dedicated receiver above the surface of the water.
An interesting application of swimming robots is the exploration of marine life, as well as the study of the biological materials buried deep under the sea.
The main purpose of the augmentation robots is to enhance the current human capability or to artificially return the ability which a human may have lost due to unfortunate circumstances.
In general, there are 3 main categories of augmenting robots:
Prosthesis: It is an artificially made limb to replace the amputated or less-developed part of a body. They are further divided into Lower Limb Prosthesis (Below Abdomen) and Upper Limb Prosthesis (Above Abdomen). Some of the most common applications include Transtibial Prosthesis (Below-Knee), Transfemoral Prosthesis (Above-Knee), and Prosthetic Hand.
Orthosis: It does not replace any limb, rather, it is used to correct or enhance the use of a body part via a supporting structure of belts and braces. These devices are very useful for increased mobility and pain reduction due to a specific body condition i.e., Toe Walking, Flatfeet, Pigeon Toes, etc.
Exoskeleton: It neither replaces nor corrects any body part, instead, it supports or amplifies the power/ lifting-capacity of a limb. It finds applications for the physio-therapy of paralyzed and old-age patients. Apart from that, industrial workers can wear an exoskeleton while Lifting Heavy Loads or doing REPETITIVE TASKS for longer periods to avoid exhaustion.
2. Types of Robots based on AUTONOMY
Robots are categorized based on AUTONOMY as per the level of assistance required from a human to perform a dedicated task.
In critical applications, the autonomy level of the robots varies as per the sophistication required during the task…
Tele-Operated robots have all the necessary components of a robot, but they cannot make independent decisions.
A human observes the sensory output of such a robot and commands its actuators to move in a specific direction or complete a specific task.
In simple words, the mechanical structure of the robot is remotely controlled by a human operator, and not by the robot itself.
A teleoperated robot is mostly used in an environment where the safety of human life is at risk i.e., extreme geographical conditions, environment exposed to radiation, and adverse weather, etc.
These robots go one step further in their autonomy as compared to Tele-Operated robots. They can move all by themselves but not without explicit instructions from a human operator.
In simple terminology, a human operator commands medium to high-level tasks while the robot automatically figures out how to achieve them.
From the operation perspective, a human monitor the output from the sensors installed within the robot and instructs the robot to head in a specific direction or move some of its actuators.
Upon receiving the instruction, robotic actuators manipulate the robotic structure to carry out the desired task.
As the name implies, these robots achieve the highest degree of autonomy. They operate independently of the human operators and possess a completely autonomous behavior. There is no need for human supervision over these robots.
In most cases, these robots are application dependent and do not operate out of their expertise. They process the sensory signals themselves and then manipulate their actuators to complete a specific task at hand.
For example, a robotic vacuum cleaner has one sole purpose of cleaning a room. Once POWERED ON, it senses the dirt on the ground, sucks it, and then moves on to the next portion of the room while avoiding all the obstacles along the way.
During the operation, the vacuum cleaner operates in a completely autonomous manner, plus no additional human supervision is required.
Applications of Robotics
If you have come this far, you are now well on your way to develop the most amazing robot in this world.
However, this is just a starting point and will only get you much far. If you want to develop a sophisticated robot, you need to Practically Implement what you have learned so far.
In simple words, you need to get your hands dirty i.e., try some sensors, control a motor, design a mechanical structure, implement a path planning algorithm, etc.
Make some things; Break some things; And you will learn how interesting is the development of a robot…
However, you should NEVER try to develop a robot that shoots aimlessly in the air. There must be an underlying problem, which your robot should try and solve!
To help you get started, I am listing a few interesting applications:
- A mini-robot to Accompany and Carry the Food/Water for every soldier during War
- A cheap Body Suit to assist a laborer during Weight Lifting
- A flying Agriculture Robot which can detect Dry Patches and Weeds in a Crop
- An Emergency Stretcher Robot to carry the patients from the ambulance into the Operation Theatre
- A Personal Assistant robot for women to assist in daily household tasks i.e., cooking, washing, cleaning, etc.
- A Miniature Security Robot which can pass undetected into a building full of criminal activities, and stream a live video of the place
- A team of Flying Swarm Robots acting as Fire Extinguishers
These are some of the ideas which I could come up with right now…
However, if you have more application-based ideas, don’t forget to share them in the comments section. Who knows…? I may help you find an investor to give your entrepreneurial idea some air to Scale and Grow!
Well, that’s it from my side! Feel free to share the article with your friends, colleagues, and business partners.
May the Force be With You!
He is the owner and founder of Embedded Robotics and a health based start-up called Nema Loss. He is very enthusiastic and passionate about Business Development, Fitness, and Technology. Read more about his struggles, and how he went from being called a Weak Electrical Engineer to founder of Embedded Robotics.