Autonomous Mobile Robots (AMR)

Definitional Boundaries & Taxonomy

Autonomous Mobile Robots (AMRs) represent a paradigm shift in intralogistics and mobile robotics. Unlike traditional Automated Guided Vehicles (AGVs) that rely on fixed infrastructure: such as magnetic tape, RFID tags, or optical lines: AMRs utilize natural feature navigation. They possess onboard computational intelligence to understand their environment dynamically, plan obstacle-free trajectories, and execute complex logistical tasks without human intervention. The defining boundary of an AMR lies in its capability for Simultaneous Localization and Mapping (SLAM) and real-time dynamic obstacle avoidance. Taxonomically, AMRs are classified by payload capacity and kinematic structure: Unit Load Carriers (for totes and small boxes, 50-200 kg), Pallet Movers/Forklifts (1000-2000+ kg), Tuggers (towing wheeled carts), and highly specialized Heavy-Duty platforms for aerospace or automotive chassis transport.

Historical Evolution

The phylogenetic tree of AMRs traces back to the late 1960s with "Shakey the Robot" at Stanford Research Institute, which pioneered the A* search algorithm for pathfinding. However, industrial mobility was dominated by wire-guided AGVs (developed by Barrett Electronics in 1953) for decades. The modern AMR era began in the late 1990s with the formalization of probabilistic robotics (Thrun, Burgard, Fox), enabling robust SLAM algorithms. The commercial watershed moment occurred in 2012 when Amazon acquired Kiva Systems, validating the decentralized, grid-based robotic fulfillment model. By the mid-2010s, the commoditization of LiDAR technology (driven by autonomous vehicle research) and the standardization of the Robot Operating System (ROS) drastically lowered the barrier to entry, catalyzing an explosion of AMR startups. Today, we are in the era of "Vision-SLAM" and AI-driven spatial intelligence.