Robotics Overview - Concepedia

Principles

Definition

Robotics is the interdisciplinary field concerned with the design, construction, operation, and application of robots. This research area investigates the integration of mechanical engineering, electrical engineering, computer science, and other disciplines to create intelligent machines capable of autonomous action and interaction with their environment.

Ontological type

Core Subfields

Control and Planning

Mechanical Design

Key developments

Key Figures

Integrated Robust Control era

Yoram Koren [1] is a leading robotics researcher shaping the Integrated Robust Control era, with affiliations at Technion – Israel Institute of Technology [3] and the University of Michigan [4]. His contributions include real-time obstacle avoidance for fast mobile robots [6] and the vector field histogram-fast obstacle avoidance for mobile robots [5], which provided real-time obstacle avoidance capabilities and scalable perception-driven planning essential for safe, autonomous operation in dynamic environments. J. Borenstein [2] is a pivotal roboticist in this era, with affiliations at Technion – Israel Institute of Technology [3] and the University of Michigan [4]. His papers The vector field histogram-fast obstacle avoidance for mobile robots [5] (1991) and Real-time obstacle avoidance for fast mobile robots [6] (1989) introduced fast, robust obstacle avoidance and integrated sensing with planning to enable real-time navigation and safe interaction in complex environments.

Probabilistic Uncertainty Management era

Andreas Geiger [1] is active in Karlsruhe Institute of Technology [3] and University of Kassel [4] during this era. He contributed through the KITTI vision benchmark suite [7], which established standard datasets and evaluation protocols that enabled robust benchmarking and reproducible progress in autonomous perception and mapping. Sebastian Thrun [2] is affiliated with Stanford University [5] and University of Pittsburgh [6] during this era. His contributions include probabilistic robotics [8], collaborative multi-robot exploration [9], and a real-time algorithm for mobile robot mapping with applications to multi-robot and 3D mapping [10], which advanced probabilistic state estimation, scalability of multi-robot sensing, and real-time mapping essential for uncertainty-aware robotics.

Data-Driven Perception-Action era

Juan D. Tardós[1] is associated with Universidad de Zaragoza[3] in this era, notably contributing to open-source SLAM through ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras[6]. These contributions provided robust multi-sensor SLAM across monocular, stereo, and RGB-D inputs, delivering crucial perception-localization capabilities that undergird data-driven perception-action pipelines in this era, and their open-source release accelerated benchmarking and integration. Sergey Levine[2] is a leading figure at Stanford University[4] and the University of Washington[5] during this era, where his work on data-driven robotics has shaped perception-action capabilities. His 2017 paper Learning hand-eye coordination for robotic grasping with deep learning and large-scale data collection[7] and his 2018 paper Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping[8] demonstrate how deep learning on large data and domain adaptation strategies propel robust grasping and sim-to-real transfer, strengthening scalable perception-action pipelines.