Autonomous mobile robot (AMR)
Also known as: autonomous mobile robots
An autonomous mobile robot (AMR) is a wheeled robot that navigates on its own, using sensors and onboard maps to move materials around warehouses, factories, hospitals, and stores.
Unlike older automated guided vehicles that follow fixed tracks or magnetic tape, AMRs plan their own routes and steer around obstacles. While AMRs have become one of the most practical forms of robotics, they still need clean floors, good Wi-Fi, and well-designed workflows to deliver their promised savings.
Autonomous mobile robot (AMR), explained at three levels
At home
Think of a delivery driver who knows the building, plans the route, and steps around people in the hall. An autonomous mobile robot does that on wheels. It first drives around to build a map with lasers and cameras, then uses that map to plan paths to wherever it's sent, watching constantly for obstacles and slowing or steering around them.
At work
An autonomous mobile robot is a self-navigating wheeled robot that moves goods in warehouses, factories, hospitals, and stores. Unlike older automated guided vehicles, it doesn't follow fixed tape or wires; it plans its own routes. Businesses use AMRs to cut walking and carrying, raise throughput, and move staff to work that needs judgment, without rebuilding the facility.
At home
An AMR carries lidar scanners near the floor, often cameras and depth sensors, wheel encoders, and an onboard computer. It locates itself on its stored map by matching what its sensors see, plans a path, and follows it while a safety scanner checks for anything in the way. If a person steps in front, it slows or stops within set zones. When the battery runs low, it drives to a dock and charges itself.
At work
AMR deployments run on fleet management software that receives tasks from a warehouse management system, assigns them to robots, and manages traffic at intersections and narrow aisles. Common models include goods-to-person picking, where robots such as Locus bots meet pickers; cart and pallet moving between work cells, as OTTO and MiR robots do; and hospital deliveries. Integration with doors, elevators, and conveyors is often the longest part of a project.
At home
AMRs avoid the fixed infrastructure of earlier robots. Changing a route means editing a map, not tearing up floor tape. The trade-off is less predictability: an AMR may take a different path each time, and crowded aisles slow it down. While they are less glamorous than humanoids, in my opinion that flexibility is exactly why AMRs have spread faster than almost any other robot.
At work
Safety rests on standards such as ISO 3691-4 for driverless industrial trucks and ANSI/A3 R15.08 for industrial mobile robots, which set requirements for scanners, speed zones, and stopping distances. Failure modes include Wi-Fi dead spots that strand robots, mixed fleets from vendors that don't coordinate, and uneven or wet floors. Interoperability standards such as VDA 5050 help. I'd measure return on investment in throughput and labour hours, and map workflows before automating them.
Real-world examples
At home
Nadia noticed a robot carrying meal trays down the hallway when visiting her grandmother in a care home. It stopped politely when a resident crossed its path.
At work
Jerome manages a distribution centre near Montreal. Introducing AMRs to carry totes between pickers and packing stations cut his team's daily walking dramatically and increased orders shipped per shift.
How it compares
| feature | Autonomous mobile robot (AMR) | Automated guided vehicle (AGV) |
|---|---|---|
| Navigation | Plans its own routes | Follows fixed paths |
| Obstacles | Goes around them | Stops and waits |
| Changing routes | Software update | Physical changes |
| Setup | Mapping the building | Installing tape, wires, or markers |
AMRs adapt to changing layouts. AGVs are simpler and predictable on fixed routes.
For analysts
AMRs can pay back quickly in the right warehouse, but I'd map workflows first. Automating a bad process just gets you a faster bad process.
History & origin
Automated guided vehicles (AGVs), which followed wires or tape, began appearing in factories in the 1950s. Kiva Systems, founded in 2003, built robots that brought shelves to warehouse workers; Amazon bought Kiva in 2012 for US$775 million, turning it into Amazon Robotics, and said it had deployed its one millionth robot in 2025. Kitchener, Ontario's Clearpath Robotics launched its OTTO Motors industrial AMRs in 2015, and Rockwell Automation acquired the company in 2023. Companies such as Locus Robotics and MiR spread AMRs to smaller warehouses and factories. In my view, AMRs are the most underrated robotics success story, because they work quietly instead of making headlines.