Introduction to Robotics: Core Concepts and System Architecture

Imagine you are sitting in a café on a busy weekday afternoon. A small wheeled robot navigates between crowded tables, adjusts its path to avoid a toddler running across the floor, and stops precisely at your table to deliver a latte. This seemingly simple interaction is the product of decades of advancement in robotics, a discipline that integrates mechanical engineering, electrical engineering, computer science, artificial intelligence, and control theory to build systems capable of perceiving their environment, making decisions, and executing physical actions autonomously. For beginners entering this interdisciplinary field, the first and most critical step is to establish a holistic understanding of what robotics is, how a typical robot system is architected, and how its core functional modules interact to enable complex real-world behaviors. This chapter lays that foundational groundwork, starting with a formal definition of robotics, tracing its historical evolution, breaking down the core components of a robot system, introducing key foundational concepts, and outlining the learning path of this book to support your journey from novice to practitioner.

1.1 Definition and Scope of Robotics

At its core, robotics is the science and engineering of designing, building, and operating robots: physical systems that interact with the physical world through sensors and actuators, and operate with varying degrees of autonomy. Unlike traditional automated systems, which are programmed to execute a fixed, repetitive sequence of actions with no ability to adapt to changing environments, robots are defined by their capacity for *environmental perception*, *decision-making*, and *adaptive execution*. A traditional industrial automation line for assembling a smartphone can only perform a fixed set of tasks if the component placement is exactly as expected; if a component is misplaced or missing, the line will halt or produce defective products. In contrast, a modern collaborative robot (cobot) equipped with vision sensors can detect misplaced components, adjust its gripping position in real time, and alert human operators to errors, enabling flexible, small-batch production.

This distinction highlights the three defining characteristics of modern robotics: 1. Physical embodiment: Robots interact directly with the physical world through mechanical structures and actuators, unlike software agents that exist only in digital environments. This embodiment introduces unique challenges such as dynamics constraints, contact forces, and sensor noise that do not exist in purely digital systems. 2. Perception-action loop: Robots operate by continuously sensing their environment, processing sensory data to estimate their own state and the state of the world around them, making decisions based on this estimated state, and executing actions that change the state of the world, which is then re-sensed to close the loop. 3. Autonomy: The level of autonomy of a robot ranges

The concept of autonomy, a defining characteristic of modern robotics, is formally standardized by the SAE International J3016 framework, which defines six levels of driving/robotic autonomy ranging from Level 0 (no automation, full human control) to Level 5 (full automation under all conditions). Translating this framework to general robotics: Level 0 systems include traditional fixed-sequence industrial automation lines with no environmental awareness; Level 1 systems, such as basic consumer扫地机器人 (robotic vacuums) equipped only with cliff and bumper sensors, can react to immediate hazards but operate with no state estimation or adaptive decision-making; Level 2 systems, including most modern automated guided vehicles (AGVs) that follow magnetic tapes or 2D barcode paths, can execute pre-defined routes and avoid static obstacles but cannot adapt to dynamic environmental changes without human intervention; Level 3 systems, such as commercial delivery robots operating in mapped office or mall environments, can perform autonomous navigation and simple task execution but will pause and request human assistance when encountering unmodeled obstacles or edge cases; Level 4 systems, including fully autonomous warehouse AMRs and last-mile delivery robots operating in geofenced urban districts, can handle all operational scenarios within their designated operational design domain (ODD) without human input; and Level 5 systems, which do not yet exist for general-purpose robotics, would be capable of operating safely in any unstructured, dynamic environment, from disaster zones to private homes, with no pre-programmed constraints. This autonomy spectrum highlights that robotics is not a binary discipline but a continuum of capability, with most deployed systems today falling between Level 1 and Level 3.

1.2 Historical Evolution of Robotics

While the term "robot" was first introduced to the public in 1920 by Czech writer Karel Čapek in his play *R.U.R. (Rossum's Universal Robots)*, derived from the Czech word *robota* meaning "forced labor" or "drudgery", the conceptual roots of robotics extend back millennia. Ancient Greek engineers built automated doors and theater props powered by water and steam as early as the 1st century CE, and Leonardo da Vinci designed a mechanical knight capable of sitting, standing, and moving its arms in the late 15th century, powered by a system of pulleys and cables. The Industrial Revolution of the 18th and 19th centuries brought the first programmable automated machines, including Joseph Jacquard's 1804 Jacquard loom, which used punched cards to control weaving patterns, a precursor to modern computer programming.

The modern era of robotics began in the mid-20th century, driven by advances in control theory, computer science, and materials engineering. In 1954, American inventor George Devol filed a patent for the first digitally operated, programmable robotic arm, a design that would become the foundation of industrial robotics. In 1961, the first commercial robot, Unimate, was installed on a General Motors assembly line in Trenton, New Jersey, where it performed the dangerous task of lifting hot die-cast metal parts from assembly lines and welding them into car bodies. Unimate could lift objects weighing up to 200 pounds (90 kg) and operated for up to 24 hours a day with a failure rate 10 times lower than human workers, immediately demonstrating the economic and safety value of robotic automation.

The 1960s and 1970s saw the emergence of mobile robotics, with the Stanford Research Institute (SRI) launching the Shakey project in 1966, which produced the first general-purpose mobile robot capable of perceiving its environment, reasoning about tasks, and navigating autonomously. Though Shakey moved at a top speed of just 2 feet per second (0.6 m/s) and required up to hours to complete simple tasks like moving a box between rooms, it pioneered core technologies including computer vision, natural language processing, and automated planning that remain central to robotics today. The 1980s saw the commercialization of mobile robots for material handling, with the first automated guided vehicles (AGVs) deployed in factories and warehouses, while the 1990s brought the first service robots, including early commercial robotic vacuum cleaners and surgical robotic prototypes.

A pivotal moment for autonomous mobile robotics came in the early 2000s with the DARPA Grand Challenge, a series of competitions organized by the U.S. Defense Advanced Research Projects Agency to accelerate the development of self-driving vehicles for military applications. The 2004 challenge, which required vehicles to navigate a 150-mile (240 km) desert route, saw no team complete the course, with the top finisher traveling just 7.4 miles (12 km) before crashing. Just one year later, in 2005, Stanford University's Stanley vehicle completed the course in 6 hours and 54 minutes, winning the $2 millio