AGV

Also known as: Automated guided vehicle

On this page 7

An AGV (automated guided vehicle) is a driverless, battery-powered industrial vehicle that transports materials along predefined routes using a guidance system such as magnetic tape, floor markers or laser reflectors. Within robotics fundamentals, the AGV is the classic form of the mobile robot for intralogistics: it follows paths defined by engineers rather than planning its own routes, which makes its behavior highly predictable, repeatable and easy to certify.

AGVs move pallets, carts, rolls, car bodies and other loads between production cells, storage areas and shipping docks. They have been used since the 1950s, and modern fleets are coordinated by a fleet management system that assigns tasks, controls traffic at intersections and schedules charging.

How an AGV navigates and moves loads

Every AGV combines four subsystems: guidance and localization, which tells the vehicle where it is relative to its path; drive and steering, which moves it; safety sensors, which stop it before a collision; and load handling, such as forks, rollers, lift tables or tow hitches. A vehicle controller links these parts and communicates with the fleet manager over Wi-Fi.

Routes are stored as a network of segments and stations. When the fleet manager assigns a transport order, the AGV drives segment by segment, reserving each section so that two vehicles never occupy the same zone. At the destination, a fine-positioning step aligns the vehicle precisely with the conveyor, rack or machine before the load is transferred.

Navigation methodHow position is determinedTypical accuracyFlexibility
Wire guidanceSenses a signal from a wire in the floorHighVery low, route changes need floor work
Magnetic tapeFollows tape or magnets on the floorRoughly a centimeterLow, tape must be relaid
Floor codes and inertialReads grid markers, dead-reckons betweenMillimeters to a centimeterMedium
Laser reflectorTriangulates from reflectors on wallsAbout a centimeter or betterHigh, routes are changed in software
Natural featuresMatches scans to a map of the buildingOne to a few centimetersHigh, no infrastructure needed
Main AGV navigation methods

Types of AGVs and typical parameters

AGVs are classified mainly by how they carry the load. The right type depends on the load carrier, the transfer points and the required throughput.

TypeTypical loadTypical application
Tugger or tow AGVTrains of cartsLine-side supply in assembly plants
Unit load AGVSingle pallet or container on top deckTransfers between conveyors
Forklift AGVPallets at floor or rack heightWarehouse storage and dock loading
Underride or lift AGVRacks or carts lifted from belowGoods-to-person and kitting
Assembly line AGVProduct being built, such as an engineFlexible assembly lines
Heavy-load AGVMany tons, such as coils or moldsSteel, paper and machine building
Common AGV types

Typical indoor AGVs travel at about 1 to 2 meters per second on open paths and much slower near people or stations. Payloads range from tens of kilograms for small carts to tens of tons for heavy-load platforms. Energy comes from lead-acid or, increasingly, lithium-ion batteries, often with opportunity charging at stations during short idle periods so that vehicles can operate around the clock.

AGV compared with an autonomous mobile robot

The most important distinction is between an AGV and an autonomous mobile robot (AMR). An AGV follows fixed, engineered paths and stops when something blocks the way; an AMR plans its own path on a map and drives around obstacles. The line has blurred because many AGVs now use natural-feature navigation, but the control philosophy still differs.

AspectAGVAutonomous mobile robot
PathPredefined route networkPlanned dynamically on a map
Obstacle responseSlows and stops, waitsReplans and drives around
PredictabilityVery highLower, routes vary
Heavy loadsCommon, up to many tonsMostly light to medium loads
CommissioningRoute engineering requiredMapping and zone setup
Best environmentStructured, repetitive flowsChanging layouts, mixed traffic
AGV and AMR compared

Predictability is not a weakness. In plants where forklifts, people and automated vehicles share aisles, knowing exactly where a vehicle will drive simplifies safety assessment and makes throughput easier to calculate.

Safety standards and system integration

AGVs operate near people, so safety is designed in from the start. Safety laser scanners monitor protective fields that grow with speed and switch with direction; if a person enters the field, the vehicle slows or performs a safety-rated stop. Bumpers, emergency stop buttons, warning lights and acoustic signals complete the system. In Europe, driverless industrial trucks are commonly assessed against ISO 3691-4, and in North America the ANSI/ITSDF B56.5 standard is widely referenced.

Integration matters as much as the vehicle. The fleet manager connects to the warehouse management or manufacturing execution system, to automatic doors, elevators and conveyors. In multi-vendor installations, the VDA 5050 interface defines a common communication protocol between vehicles and a master control, which reduces dependence on one supplier.

Common pitfalls in AGV projects

Most AGV problems come from the environment and the process rather than from the vehicles themselves.

  • Poor floor condition: cracks, slopes and joints affect wheels, guidance and load stability.
  • Blocked paths: pallets left in aisles stop AGVs repeatedly, because they wait instead of driving around.
  • Underestimated traffic: intersections and narrow aisles become bottlenecks when the fleet grows.
  • Wi-Fi gaps: dead zones delay orders and can stop vehicles at critical points.
  • Inconsistent load carriers: damaged or nonstandard pallets cause failed pickups at stations.

A simulation of the material flow before purchase, with realistic order peaks and charging time, is the most reliable way to size the fleet and to find bottlenecks while the layout can still be changed cheaply.

In plain words

An AGV is a driverless cart or forklift that follows invisible rails through a factory or warehouse. The rails might be a strip of tape on the floor, small codes, or reflectors on the walls that the vehicle uses to find its position.

Because it always takes the same routes, everyone knows where it will go. If something is in the way, it simply stops and waits instead of improvising, which makes it safe and dependable for moving heavy goods all day.

A real-world example

An automotive supplier assembles seats on two production lines and needs a steady flow of parts from a central storage area 150 meters away. The company installs six tugger AGVs, each pulling a train of three carts, on a fixed loop with laser reflector navigation and opportunity charging at the loading station.

The fleet manager receives part requests from the manufacturing execution system and dispatches trains every few minutes. Safety scanners slow the vehicles in the two aisles shared with pedestrians. Manual forklift trips to the lines drop to a few exceptions per shift, line stops caused by missing parts become rare, and the predictable delivery rhythm lets the plant reduce the buffer stock kept at each workstation.