Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.
A robotic prosthetic hand is a close replica of an actual human hand
UW News (CC BY 2.0) https://creativecommons.org/licenses/by/2.0/

What makes a robot a robot?

Most people are familiar with the idea of a robot. But what makes a robot a robot, and not just another machine? In this step, we define some of the key characteristics of a robotic system.

What is a robot?

A key feature of robots is that they interact with the world, making changes to the world through their actions and responding to events in the world. Robots perform useful tasks, extend the capabilities of humans and reduce our risks when operating in hazardous environments.

Robotic systems can be defined as interconnected, interactive, cognitive and physical tools that are able to:

  • perceive the environment using sensors
  • reason about events
  • make plans using algorithms implemented in computer programs
  • perform actions enabled by actuators

This ‘feedback loop’ of autonomous sensing, perception, cognition and action is what distinguishes the robot from other machines (see figure 1).

A diagram demonstrating a RAS feedback loop. A robot senses and perceives the world, which feeds into it's cognition, influencing it's actions which then have implications in the world. A robot or autonomous system consists of the entity itself and its interaction with the world including other RAS and humans. The key elements of any RAS are the ability to sense and perceive the world, make intelligent decisions and take appropriate actions.

Example: a driverless car

The driverless car is an example of a typical robotic system:

  • Sensors: the car has a number of sensors that allow it to perform autonomously, including LIDAR (light detection and ranging), video cameras, RADAR (radio detection and ranging), wheel encoders (which measure wheel rotation to estimate distance travelled) and GPS (global positioning system) which is used to measure the car’s location in the world.

  • Perception: perception algorithms transform the raw data into labelled objects e.g. other cars, pedestrians and road signs.

  • Cognition: cognition algorithms allow the car to plan actions based on current perceptions and goals (i.e. desired destination and route, avoiding obstacles).

  • Action: the actions are implemented by low-level control algorithms that manipulate the steering wheel, accelerometer and brake to move the car, and which take account of the vehicle dynamics, road surface and environmental conditions.

A photo showing an autonomous car. The video camera on the windshield is labelled with the text: "Positioned near rear-view mirror, spots moving objects (such as pedestrians) as well as detecting traffic light signals." The rotating sensor on the roof is labelled with the text: "Mounted on roof, carries out laser sweep of surroundings to generate 3D map of car's environment." The radar sensors on the front and rear bumpers are labelled: "Spread across front and rear of car, measures distance to obstacles in front and behind, triggering reduction in speed where necessary." The position estimator on a wheel is labelled: "Attached to the left rear wheel, tracks every minor movement of vehicle to pinpoint its exact location on computerised map." A driverless car with labelled sensors. Photo: John Greenfield (CC BY 2.0)

Key challenges to solve for the future

For robots to make an impact on human society in future, they need to be able to operate safely and effectively amongst humans. Most conventional robots operate successfully in structured environments such as factories. These are controlled spaces where the general public are not allowed to venture, or only under supervision.

A key challenge for future robots therefore, to bring them out of the factories, is to make them more robust to unstructured environments, where the environment is unknown and unexpected events can occur.

The challenges for robots in unstructured environments extend across all the robotic elements of sensing, perception, cognition and action.

In each week of this course, we’ll focus on a particular area related to current research challenges that underpin the future of robotics.

  • The focus of Week 1 is on sensing and perception, with an introduction to cognition and action using planning techniques such as primitive actions.
  • The focus of Week 2 is on bioinspired robotics, which links to cognition and aims to use design principles from the natural world to improve adaptability, versatility and robustness.
  • The focus of Week 3 is on cooperative robotics, where robots combine in teams to solve problems collectively, and expands into ethical issues in robotics, as well as skills a robotics engineer will need in the future, such as computer programming.

Discussion

Why is an automated dishwashing machine not a robot? (Or is it?)

Share this article:

This article is from the free online course:

Building a Future with Robots

The University of Sheffield

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join:

Contact FutureLearn for Support