The Robotics WEBook

An online textbook about robots and other mechatronic systems

Robotics

(Figures)

Roboticists develop mechanical devices that can move by themselves, whose motion is be modelled, planned, sensed, actuated and controlled, and whose motion behaviour can be influenced by the programmed task as well as by the environment in which the robot device operates. Robots are called “intelligent” if they succeed in moving in safe interaction with an unstructured environment, while autonomously achieving their specified tasks.

This definition implies that a device can only be called a “robot” if it contains a movable mechanism, influenced by sensing, planning, actuation and control components. It does not imply that a minimum number of these components must be implemented in software, or be changeable by the “consumer” who uses the device; for example, the motion behaviour can have been hard-wired into the device by the manufacturer.

The presented definition, as well as the rest of the material in this part of the WEBook, cover not just “pure” robotics or only “intelligent” robots, but rather the somewhat broader domain of robotics and automation. Automation is then identified with the mature, “low-level” end of robotics. So, the scope of the WEBook includes “dumb” robots such as: metal and wood working machines, “intelligent” washing machines, industrial dish washers and pool cleaning robots, etc. These examples all have moving parts, and they incorporate sensing, planning and control, but often not in individually separated components. For example, the sensing and planning behaviour of the pool cleaning robot have been integrated into the mechanical design of the device, by the intelligence of the human developer.

This part of the WEBook describes, in a consistent and motivated structure, the core science and technology of robotics. Nevertheless, the chosen structure and emphasis represent only one of the many possible “views” that one can want to have on the robotics domain. (A later phase in the WEBook development could support the creation of such different “semantic views” on the WEBook material.)

In the same vein, the above-mentioned “definition” of robotics is not meant to be definitive or final. It is only used as a framework to structure the various chapters of the WEBook.

Much of the knowledge and science used in robotics are shared with other domains, such as physics and mathematics, systems and control, computer science, virtual reality and character animation, machine design, computer vision, artificial intelligence, cognitive science, biomechanics, etc. In addition, “robotics” ideas, concepts and algorithms are being applied in an ever increasing number of “external” applications.

The first WEBook ambition is to provide textbook descriptions of the components of robotics agents; for example, kinematics and dynamics of mechanical systems; robot localization and motion planning algorithms; force and impedance control. A second focus is the “ontological” classification of robotic systems and domains. The more glossary-, encyclopedia-like descriptions of robotics terms and concepts can probably be provided better by the Wikipedia. In all three of these content description levels, the reader should be aware of the unspoken background sensitivity that can creep into, both, the presentation by the content creators and the content interpretation by the readers. One common case of the occurrence of such implicit background assumptions shows up in the semantics of the various research and application domains (“medical robotics”, “industrial robotics”, “field robotics”, …): each of these domains has lots of hidden assumptions about the relevant scales and contexts that are made in the design of the robotics agents designed to operate in that domain.

Robotic agents

This figure depicts the components that are part of all robotic “agents”. (The terms “agent” or “device” or “robot” are loosely used as synonyms, all referring to the same man-made thing.) The robot agent is some mechanical device (“mechanism”) that moves around in the environment, and, in doing so, physically interacts with this environment. This interaction involves the exchange of physical energy, in some form or another. Both the robot mechanism and the environment can be the “cause” of the physical interaction through “Actuation”, or experience the “effect” of the interaction, which can be measured through “Sensing”.

Robotics as an integrated system of control interacting with the
physical world.
Robotics as an integrated system of control interacting with the physical world.

Sensing and actuation are the physical ports through which the “Controller” of the robot determines the interaction of its mechanical body with the physical world. As mentioned already before, the controller can, in one extreme, consist of software only, but in the other extreme everything can also be implemented in hardware.

In many practical cases, “people” (i.e., papers, software libraries, designs, …) focus only on an agent-centric view, i.e., without taking the environment properties into account. In addition, they make lots of implicit scale and context assumptions.

Parts of robotic agent
Agent-centric view on robotics.

Robotic systems and domains

Robotics is, to a very large extent, all about system integration. While the previous Section focuses on the robot as a stand-alone man-made Agent, this Section explains the complementary focus, that of positioning the robot agent as one part in a “robotic system” that includes the real world. The goal of a particular system is to achieve a particular Task in a particular Environment (figure). So, the system designers make a particular choice and trade-off between which parts of the system are taken care of by the Agent, or by the Task, or by the Environment. For example, a robotic assembly system can improve its performance by introducing more structure in the environment (chamfers, tools customized to each single task, etc.), or more complex mechanisms in the agent, or more customized task descriptions.

Agent-Task-Environment triangle
The Agent-Task-Environment continuum of robot systems: every real robotics system is situated somewhere within the boundaries of this triangle.

Some continous regions within the Agent-Task-Environment triangle are identified by means of a particular name. Such a robotics domain is defined (or, rather, known to the practitioners in the field, without explicit definition) as the set of all robotic systems that share most of their design trade-offs. As in the agent-centric view, this includes lots of implicit scale and context assumptions.

Scales in robotic agents and systems

The above-mentioned “agent” and “system” descriptions are complemented by a “scale” description, i.e., the following system scales have a large influence on the specific content of the sensing, actuation, mechanism and control components in a robotic agent, and on the design trade-offs in a robotic system.

Obviously, these scale parameters never apply completely independently to the same system. For example, a system that must react at microseconds time scale can not be of macro mechanical scale and involve a high number of communication interactions between subsystems.

Background sensitivity

No description of science or technology is ever fully objective or context-free: it is very difficult for content creators to “forget” their background when writing their contributions. One of the most visible of such background sensitivities comes from the historical evolution that gave robotics has, roughly speaking, two faces:

  1. The mathematical and engineering face, which is quite “standardized” in the sense that a large consensus exists about the tools and theories to use (“systems theory”).
  2. The AI face, which is rather poorly standardized, not because of a lack of interest or research efforts, but because of the inherent complexity of the science and technology of “intelligent behaviour.”

Research in engineering robotics follows the bottom-up approach: existing and working systems are extended and made more versatile. Research in artificial intelligence robotics is top-down: assuming that a set of low-level primitives is available, how could one apply them in order to increase the “intelligence” of a system. The border between both approaches shifts continuously, as more and more “intelligence” is cast into algorithmic, system-theoretic form. For example, the response of a robot to sensor input was considered “intelligent behaviour” in the late seventies and even early eighties. Hence, it belonged to A.I. Later it was shown that many sensor-based tasks such as surface following or visual tracking could be formulated as control problems with algorithmic solutions. From then on, they did not belong to A.I. any more.

The terminology and systems-thinking of both backgrounds are significantly different, hence the WEBook will contain sections on the same material but written from various perspectives. This is not a “bug”, but a feature: having the different views in the same WEBook can only lead to a better mutual understanding and respect.