Robotics Foundations and System Architecture

Introduction

Robotics is the engineering discipline concerned with machines that perceive their environment, make decisions, and produce physical action. It combines mechanics, electrical and electronic engineering, control, computing, sensing, artificial intelligence, and human factors. A robot is not merely a programmable machine. In the engineering sense, it is an integrated system whose mechanical structure, actuators, sensors, controller, software, power system, and end effector work together to achieve a task.

Robotics and kinematics begin with a deceptively simple question: how can a mechanism be designed and controlled so that its tools occupy desired positions and orientations? Answering this question requires a precise vocabulary and an understanding of system architecture. This chapter establishes that foundation by examining robot classifications, mechanical structures, joints and links, degrees of freedom, actuation, sensing, control, workspaces, and open- and closed-chain mechanisms.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Define fundamental robotics and kinematics terminology.
2. Distinguish industrial, mobile, collaborative, service, medical, and autonomous robots.
3. Identify the functions of manipulators, links, joints, actuators, sensors, controllers, and end effectors.
4. Explain degrees of freedom and workspace using appropriate engineering language.
5. Compare open-chain and closed-chain architectures.
6. Select an appropriate robot architecture for a practical application.
7. Recognize safety, ethical, and professional responsibilities in robot deployment.

1. What Is a Robot?

The word robot is used broadly, but engineering definitions usually emphasize three properties: physical embodiment, programmable behavior, and the ability to perform actions in the physical world. The International Federation of Robotics commonly distinguishes industrial robots from service robots according to their application and degree of interaction with people. Standards such as ISO 8373 provide terminology for robots and robotic devices, while safety standards such as ISO 10218 and ISO/TS 15066 address industrial and collaborative robot operation.

A robot generally contains:

• a mechanical structure that supports and guides motion;
• actuators that generate force or torque;
• sensors that measure internal states or external conditions;
• a controller that computes commands;
• software that implements motion, sequencing, perception, and supervision;
• a power system; and
• an end effector or functional device that interacts with the task.

A robot should be distinguished from a conventional automated machine. A fixed-purpose machine may execute a repetitive sequence using cams, switches, and dedicated tooling. A robot is usually more flexible: its motion and task logic can be reprogrammed, and it often has multiple controlled axes. The boundary is not absolute, however. Some automated machines are robot-like, and some robots are highly specialized.

The term robotics system includes the robot and its wider operational environment: fixtures, conveyors, safety devices, communication networks, human operators, production software, and the workpiece. This systems perspective is essential because many failures arise not from the manipulator itself but from poor integration.

2. Major Robot Classifications

No single classification captures every robot. Robots may be classified by application, mobility, mechanical architecture, autonomy, or interaction with humans. The following categories are especially useful in engineering practice.

2.1 Industrial Robots

Industrial robots are programmable manipulators used in manufacturing and related production activities. Typical tasks include welding, painting, assembly, material handling, palletizing, machine tending, inspection, and additive manufacturing. They are often mounted on a fixed base and designed for repeatability, speed, payload capacity, and operation in structured environments.

Common industrial forms include articulated arms, SCARA robots, Cartesian robots, cylindrical robots, and delta robots. An industrial robot installation also includes tooling, fixtures, safety fencing or monitored access, programming equipment, and process equipment. The robot arm alone is not the complete production system.

2.2 Mobile Robots

Mobile robots change their location through wheels, tracks, legs, propellers, or other locomotion systems. An autonomous mobile robot (AMR) may use sensors and software to map an environment, localize itself, plan paths, and avoid obstacles. An automated guided vehicle (AGV) generally follows predefined guidance such as magnetic tape, reflectors, or mapped routes, although modern systems may combine both approaches.

Mobile robots are characterized by locomotion constraints. A wheeled platform may be unable to move directly sideways, whereas an omnidirectional platform can translate laterally. These constraints affect navigation, control, and the reachable region of the vehicle.

2.3 Collaborative Robots

A collaborative robot, or cobot, is designed to operate in a workspace where it may interact with people under specified safety conditions. Collaboration is not simply a physical property of a robot arm. It depends on the complete application, including speed, force, tooling, workpiece hazards, risk assessment, safeguarding, and validated operating modes.

Collaborative operation may involve hand guiding, safety-rated monitored stop, speed and separation monitoring, or power and force limiting. A robot marketed as collaborative does not make every task safe. Sharp tools, hot parts, crushing points, and unexpected motion remain hazards.

2.4 Service Robots

Service robots perform useful tasks for people or organizations outside traditional industrial manufacturing. Examples include domestic cleaning robots, agricultural robots, logistics robots, hotel delivery robots, inspection robots, and public-space information robots. Service robots often encounter variable environments and untrained users, so perception, navigation, human–robot interaction, and fault recovery become especially important.

2.5 Medical Robots

Medical robots support diagnosis, therapy, rehabilitation, surgery, prosthetics, and hospital logistics. Surgical systems may provide precise instrument manipulation, motion scaling, tremor filtering, or minimally invasive access. Rehabilitation robots may guide limbs or provide controlled resistance. Medical robots operate in a highly regulated environment, where clinical evidence, cybersecurity, traceability, human oversight, and risk management are as important as mechanical performance.

2.6 Autonomous Robots

An autonomous robot senses its environment and selects actions with limited direct human control. Autonomy is not binary; it exists on a spectrum. A robot may be autonomous in navigation but not in task selection, or autonomous during normal operation but require human intervention in unusual conditions.

A useful functional chain is:

sense → estimate → plan → act → monitor

Sensors provide data; estimation converts data into a state representation; planning selects a course of action; actuation produces motion; monitoring checks whether the result is safe and successful. Autonomy requires all these functions to be integrated, not merely the addition of an artificial-intelligence algorithm.

3. The Robotic System Architecture

A robot can be understood as a hierarchy of interacting subsystems. At the lowest level, actuators and feedback controllers regulate joint motion. At higher levels, a task planner sequences operations and coordinates perception, motion, and safety.

Figure: Functional architecture of a feedback-controlled robotic system.

3.1 Manipulator

A manipulator is a mechanical device composed of connected links and joints that supports and moves an end effector. In common industrial usage, the manipulator refers primarily to the arm mechanism, while the robot system includes the controller, drives, sensors, and other equipment.

3.2 Links and Joints

A link is a comparatively rigid body in a mechanism. Links transmit motion and forces between joints. The base is often treated as a fixed link, while the terminal link carries the wrist or end effector.

A joint connects two links and permits specified relative motion. The most common idealized joints are:

• Revolute joint (R): permits rotation about an axis.
• Prismatic joint (P): permits translation along an axis.
• Helical or screw joint: couples rotation and translation with a fixed pitch.
• Spherical joint: permits three-dimensional rotation about a common point.

Real robot joints may include bearings, gearboxes, brakes, encoders, seals, cables, and structural housings. The ideal joint model is useful for kinematic analysis, but compliance, backlash, friction, and flexibility may become important in precision applications.

3.3 Actuators and Drives

An actuator converts supplied energy into mechanical motion. Electric servo motors are common because they offer controllability, efficiency, and clean operation. Hydraulic actuators provide high force density and are used in heavy-duty or outdoor applications. Pneumatic actuators are simple and fast but compressibility can make precise positioning difficult.

The drive system includes power electronics and control interfaces. A motor may be capable of high speed, but a gearbox can increase output torque while reducing output speed. In an ideal gearbox, neglecting losses, power is approximately conserved:

where P is mechanical power in watts, τ is torque in newton-metres, and ω is angular velocity in radians per second. The relationship applies instantaneously to rotational mechanical power. With a gearbox and losses, output power is less than input power, and efficiency must be considered.

3.4 Sensors

Sensors measure variables needed for control, estimation, safety, and interaction. Proprioceptive sensors measure the robot’s internal state, such as joint position, velocity, motor current, temperature, or torque. Exteroceptive sensors measure the environment, such as cameras, lidar, radar, ultrasonic sensors, force–torque sensors, and tactile arrays.

No sensor is perfect. Measurements may contain noise, bias, delay, quantization, occlusion, or drift. Sensor selection therefore depends on range, resolution, update rate, environmental robustness, calibration requirements, and safety consequences.

3.5 Controller and Software

The controller compares desired behavior with measured behavior and generates actuator commands. A simple position feedback error can be represented as:

where q_d(t) is desired joint position, q(t) is measured joint position, and e(t) is position error, all in radians for a revolute joint or metres for a prismatic joint. This expression defines error; it is not by itself a complete control law.

Controllers may operate at several levels. A servo loop regulates motor position, velocity, or torque. A motion controller generates coordinated joint trajectories. A task-level controller manages actions such as “pick,” “place,” or “inspect.” Supervisory software handles modes, faults, communications, and safety interlocks.

3.6 End Effectors

An end effector is the device attached to the robot wrist or terminal link that performs the task. Examples include parallel grippers, vacuum cups, magnetic grippers, welding torches, spray guns, screwdrivers, milling spindles, cameras, and surgical instruments.

End-effector design is often the decisive factor in system success. It must match the workpiece geometry, mass, surface, fragility, temperature, and process requirements. Tool changing can increase flexibility but introduces docking, sensing, calibration, and failure requirements.

4. Degrees of Freedom and Configuration

A degree of freedom (DOF) is an independent coordinate required to specify a system configuration. In three-dimensional space, a freely moving rigid body has six degrees of freedom: three translations and three rotations. A robot’s joint variables provide its joint space description, often written as:

where q_i is the coordinate of joint i and n is the number of independently actuated or constrained joint coordinates in the model.

For a serial manipulator with six independent revolute or prismatic joints, the joint configuration has six DOF. This often permits independent control of three-dimensional position and three-dimensional orientation of the tool. However, six joints do not guarantee six useful task DOF: joint axes may be parallel or may reach a singular configuration.

Degrees of freedom must be distinguished from the number of motors. A passive mechanism may have mobility without an actuator at every joint; conversely, a mechanically redundant robot may have more joint coordinates than the task requires. Redundancy can improve obstacle avoidance and posture selection but increases planning and control complexity.

5. Workspace, Reachability, and Dexterity

The workspace is the set of positions or poses that the end effector can attain under specified joint limits, geometric constraints, and environmental assumptions. Important distinctions include:

• Reachable workspace: positions attainable by at least one configuration.
• Dexterous workspace: positions at which specified orientations can also be attained.
• Collision-free workspace: reachable configurations that do not collide with the robot, environment, or people.
• Task workspace: the subset actually needed for a particular operation.

Workspace is not simply the largest distance from the base. It depends on link lengths, joint limits, joint-axis arrangement, singularities, tool geometry, payload, and installation layout. A robot may reach a point but be unable to orient its tool appropriately or produce sufficient force there.

Worked Example: Planar Two-Link Workspace

Consider an ideal planar arm with two revolute joints and link lengths L₁ = 0.40 m and L₂ = 0.30 m. If both links align in the same direction, the maximum distance from the base is:

If the links fold back along one another, the minimum radial distance is:

Under the ideal assumption of unrestricted full rotation and no obstacles, the radial reach therefore lies between 0.10 m and 0.70 m. In a real robot, joint limits and collision constraints may remove large portions of this annular region. The example illustrates reachability only; it does not establish orientation capability, speed, payload, or accuracy.

6. Open-Chain and Closed-Chain Architectures

6.1 Open-Chain or Serial Mechanisms

An open-chain mechanism has a single sequence of links from the base to the end effector. Each joint contributes motion along the chain. Articulated industrial arms and many mobile manipulator arms are serial mechanisms.

Advantages include large workspace, straightforward modular construction, and flexible positioning. Disadvantages include accumulated compliance and error, because loads are transmitted through a sequence of joints. The distal links also carry the masses of downstream links, which can reduce payload-to-weight ratio and dynamic performance.

6.2 Closed-Chain or Parallel Mechanisms

A closed-chain mechanism contains one or more loops of links. In a parallel robot, several kinematic chains connect the base to a common moving platform. Delta robots and Stewart platforms are important examples.

Parallel mechanisms can provide high stiffness, low moving mass, and high speed. Their drawbacks include a more restricted workspace, complex singularity behavior, difficult calibration, and possible internal forces. A Stewart platform, for example, can control six platform motions using six variable-length legs, but its motion is governed by coupled geometry rather than independent serial joints.

6.3 Hybrid Architectures

Some systems combine serial and parallel structures. A serial arm may carry a parallel wrist, or a mobile base may carry a serial manipulator. Hybrid architectures seek a balance among workspace, stiffness, dexterity, payload, and cost.

7. Kinematic Structure and Configuration Spaces

Kinematics describes motion without directly considering forces and torques. Forward kinematics determines the end-effector pose from joint variables. Inverse kinematics determines joint variables that produce a desired pose. A robot may have zero, one, or many inverse-kinematic solutions, or the desired pose may be unreachable.

A configuration is a complete specification of the robot’s joint coordinates. The set of all allowable configurations is the configuration space, often called C-space. Obstacles in physical space become forbidden regions in configuration space. This transformation is important for collision-free path planning: a path is safe only if the corresponding configuration trajectory avoids forbidden configurations.

A singularity is a configuration at which the robot loses one or more instantaneous motion directions or gains unusually high sensitivity. Near a singularity, small Cartesian motions may require very large joint velocities, and force transmission may become unfavorable. Singularities are therefore relevant to planning, control, accuracy, and safety.

8. Feedback, Accuracy, Repeatability, and Performance

Accuracy is the closeness of the achieved position or orientation to the commanded or nominal value. Repeatability is the closeness of repeated outcomes to one another. A robot may be highly repeatable but inaccurate because of calibration error. This distinction matters in manufacturing: a process may be corrected with calibration if repeatability is good, but random variation is more difficult to eliminate.

Other performance measures include payload, speed, cycle time, resolution, stiffness, energy consumption, reliability, availability, and mean time between failures. These measures interact. Increasing speed may increase vibration; increasing payload may reduce acceleration; improving precision may increase cost and calibration effort.

9. Applied Case: Robotic Assembly Cell

Background

Consider an illustrative case involving a small manufacturer assembling electronic housings. The existing process requires an operator to pick a housing, place it in a fixture, insert four screws, and transfer the completed unit to a test station. The cycle is repetitive, and operators report wrist fatigue.

Problem

The company proposes a six-axis collaborative robot. The design team must decide whether the robot is suitable and how the complete system should be engineered.

Analysis

The task requires several capabilities: a gripper for the housing, a screwdriver or tool changer, positional accuracy at the screw locations, sensing to confirm part presence and screw seating, and safe coexistence with operators. A six-axis arm provides sufficient pose freedom, but DOF alone does not prove suitability. The workspace must include the parts rack, fixture, screw feeder, and test station. The payload must include the housing, gripper, screwdriver, and any dynamic loads.

The team identifies hazards including pinch points, dropped parts, screwdriver reaction torque, sharp edges, unexpected restart, and interaction with an operator. A collaborative designation cannot replace a task-specific risk assessment.

Approach and Solution

The proposed cell uses a compliant gripper, a wrist force–torque sensor, a presence sensor in the fixture, and a torque-controlled screwdriver. The controller uses a structured sequence:

1. verify that the cell is in a safe operating mode;
2. detect and grasp the housing;
3. confirm grasp force and part presence;
4. place the housing in the fixture;
5. verify seating;
6. insert and tighten each screw with torque monitoring;
7. transfer the assembly to test;
8. report the result and recover safely from faults.

The robot operates at reduced speed when a person enters a monitored collaboration zone. A safety-rated stop is used when the operator performs a manual intervention. The cell includes physical design features that prevent access to hazardous tooling during operation.

Outcome

In this illustrative case, the robot improves consistency and reduces repetitive handling, but it does not eliminate human work. Operators load materials, respond to faults, perform quality checks, and maintain the equipment. The final design is successful only after cycle-time trials, validation of torque and placement quality, training, and documented risk controls.

Learning Points

The case demonstrates that robot selection is a systems-engineering decision. The arm’s brand and nominal payload are less important than the combined performance of the mechanism, tooling, sensors, controller, safety system, fixture, and process. It also demonstrates that collaborative operation is an application property requiring risk assessment and validation.

10. Safety, Ethics, and Governance

Robots can injure people through impact, crushing, cutting, entanglement, heat, stored energy, or unexpected restart. Safe design follows a hierarchy: eliminate hazards where possible, reduce inherent energy and speed, use guards and protective devices, implement safety-rated control functions, and provide procedures, training, and personal protective equipment.

Relevant standards vary by jurisdiction and application, but professional practice commonly involves machinery risk assessment, industrial robot safety standards, electrical safety requirements, functional safety, electromagnetic compatibility, and medical-device regulation where applicable. Engineers must consult current legally applicable standards rather than relying on informal rules.

Ethical issues include worker displacement, surveillance, allocation of responsibility after an autonomous failure, unequal access to automation, and the use of robots in sensitive environments. Data collected by cameras and microphones may create privacy risks. Autonomous systems should provide appropriate human oversight, event logging, access control, cybersecurity protection, and clearly defined operational limits.

Artificial intelligence can improve perception, prediction, and planning, but learned models may fail under distribution shift, sensor degradation, or unusual human behavior. AI should therefore be bounded by safety monitors, validated operating domains, fallback behavior, and human authority to stop the system.

11. Contemporary Developments and Best Practices

Current robotics development emphasizes flexible manufacturing, digital twins, cloud-connected fleets, force-controlled manipulation, soft robotics, legged mobility, human–robot collaboration, and data-driven maintenance. These developments increase capability but also increase integration and cybersecurity demands.

Best practices for a robotics project include:

• begin with a measurable task specification;
• model workspace, payload, reach, speed, and tooling before purchase;
• distinguish accuracy from repeatability;
• include sensors and fixtures in the design, not as afterthoughts;
• validate all reachable and collision-free motions;
• conduct a documented risk assessment;
• design for fault detection and safe recovery;
• calibrate the robot and verify performance using appropriate tests;
• protect networks, credentials, programs, and production data;
• retain human oversight where consequences of failure are significant; and
• document assumptions, limitations, maintenance, and change control.

Chapter Summary

Robotics is an interdisciplinary field involving mechanical structures, actuation, sensing, control, computation, and interaction with the physical world. Robots may be classified as industrial, mobile, collaborative, service, medical, or autonomous, although these categories can overlap. A robotic system includes a manipulator, links, joints, actuators, sensors, controller, software, power system, and end effector.

Degrees of freedom describe independent configuration coordinates, while workspace describes attainable positions or poses subject to geometry, joint limits, and environmental constraints. Serial mechanisms use an open chain and commonly provide large, flexible workspaces; parallel mechanisms use closed chains and often provide high stiffness and speed at the cost of coupled kinematics and restricted workspace. Kinematics concerns motion, including forward and inverse relationships, configurations, and singularities.

Successful deployment requires more than selecting an arm. Tooling, sensing, fixtures, safety systems, calibration, software, human factors, cybersecurity, and regulatory obligations must be engineered as one system. Robotics professionals must therefore combine theoretical understanding with practical validation and ethical responsibility.

Key Terms

Robot: A programmable physical system capable of sensing, computation, and action.

Robotics system: The robot plus its tooling, environment, control, safety, and integration components.

Manipulator: A mechanism that positions and orients an end effector.

Link: A relatively rigid body in a mechanism.

Joint: A connection that permits specified relative motion between links.

Revolute joint: A joint providing rotation about an axis.

Prismatic joint: A joint providing translation along an axis.

Actuator: A device that converts energy into mechanical motion.

Sensor: A device that measures an internal state or environmental variable.

Controller: Hardware and software that computes commands for robot behavior.

End effector: A tool attached to the terminal link for performing a task.

Degree of freedom: An independent coordinate needed to specify configuration.

Joint space: The space of robot joint coordinates.

Workspace: The set of positions or poses attainable by the end effector.

Serial mechanism: An open-chain mechanism with a sequence of links and joints.

Parallel mechanism: A closed-chain mechanism with multiple chains supporting a platform.

Configuration: A complete specification of the robot’s relevant coordinates.

Singularity: A configuration involving loss or unusual amplification of instantaneous motion capability.

Accuracy: Closeness of an outcome to its desired or nominal value.

Repeatability: Closeness of repeated outcomes to one another.

Review Questions

Short Answer Questions

1. What distinguishes a robot system from a robot manipulator?
2. Define a revolute joint and a prismatic joint.
3. What is the difference between accuracy and repeatability?
4. What is an end effector, and why is its design important?
5. Define reachable workspace and dexterous workspace.

Long Answer Questions

1. Explain the architecture of a feedback-controlled robotic system, from sensing and estimation to planning and actuation.
2. Compare industrial, mobile, collaborative, service, medical, and autonomous robots, giving one application for each.
3. Explain the advantages and limitations of serial and parallel robot architectures.
4. Discuss degrees of freedom, joint space, workspace, and singularities in a six-axis manipulator.
5. Explain why collaborative robot safety depends on the complete application rather than on the robot label alone.

Analytical / Application Questions

1. A packaging task requires a robot to reach three stations, grasp fragile products, and operate beside a worker. Identify the required mechanical, sensing, control, and safety features, and justify your choices.
2. A serial arm can reach a point but cannot orient its welding torch correctly. Analyze possible causes using the concepts of workspace, degrees of freedom, joint limits, and singularities.
3. Compare a delta robot and a six-axis articulated robot for high-speed sorting of lightweight objects. Recommend one architecture and state the conditions that could reverse your recommendation.

Multiple Choice Questions with Answers

Question 1

Which component converts supplied energy into mechanical motion?

A. Sensor

B. Actuator

C. Controller

D. Fixture

Correct Answer: B. Actuator

Question 2

A joint that permits rotation about an axis is a:

A. Prismatic joint

B. Spherical joint

C. Revolute joint

D. Fixed joint

Correct Answer: C. Revolute joint

Question 3

Which statement best describes repeatability?

A. Closeness to the commanded value

B. Closeness of repeated outcomes to one another

C. Maximum payload under static conditions

D. The number of robot joints

Correct Answer: B. Closeness of repeated outcomes to one another

Question 4

A mechanism with one sequence of links from base to end effector is a:

A. Closed-chain mechanism

B. Parallel mechanism

C. Serial mechanism

D. Spherical mechanism

Correct Answer: C. Serial mechanism

Question 5

What does an end effector primarily do?

A. Supply electrical power to the controller

B. Perform the task at the robot’s terminal link

C. Define the robot’s software architecture

D. Measure motor temperature only

Correct Answer: B. Perform the task at the robot’s terminal link

Question 6

A freely moving rigid body in three-dimensional space has how many degrees of freedom?

A. 3

B. 4

C. 6

D. 9

Correct Answer: C. 6

Question 7

Which feature is commonly associated with parallel mechanisms?

A. High stiffness and coupled motion

B. Unlimited workspace

C. No calibration requirements

D. Only one supporting kinematic chain

Correct Answer: A. High stiffness and coupled motion

Question 8

Which sensor is exteroceptive?

A. Joint encoder

B. Motor current sensor

C. Gearbox temperature sensor

D. Lidar measuring surrounding obstacles

Correct Answer: D. Lidar measuring surrounding obstacles

Question 9

For two planar links of lengths 0.40 m and 0.30 m, the maximum straight-line reach under unrestricted alignment is:

A. 0.10 m

B. 0.30 m

C. 0.40 m

D. 0.70 m

Correct Answer: D. 0.70 m

Question 10

Why does a collaborative robot designation not automatically make every application safe?

A. Collaborative robots cannot use sensors

B. Safety depends on tooling, speed, hazards, layout, and validated risk controls

C. Collaborative robots have no controllers

D. They are limited to medical applications

Correct Answer: B. Safety depends on tooling, speed, hazards, layout, and validated risk controls