A mobile robot is a robot whose whole body moves through its environment, using wheels, tracks, legs, propellers or thrusters, rather than working from a fixed base. In robotics fundamentals it is one of the two great families of machines, set against the fixed-base manipulator such as the industrial robotic arm bolted to a factory floor.
Mobility changes the engineering problem completely. A mobile robot must know where it is, decide where to go and move safely among obstacles and people, all while carrying its own energy supply. The category ranges from a simple line-following cart to a fully autonomous mobile robot, a legged quadruped or an aerial drone.
What defines a mobile robot in robotics
The defining property in robotics is locomotion of the entire platform: the robot changes its own position and orientation in the world, which engineers call its pose. A fixed manipulator works inside a bounded workspace that never moves, while a mobile robot carries its workspace with it and can operate anywhere its locomotion system can reach.
Every mobile robot combines the same functional layers, even when they are very simple. Engineers usually describe them as a chain from sensing to action, and each layer can range from a few lines of logic to a large software stack.
| Layer | Purpose | Typical components |
|---|---|---|
| Locomotion | Moves the platform | Wheels, tracks, legs, rotors, motor drivers |
| Perception | Senses surroundings | LiDAR, cameras, ultrasonic sensors, bumpers |
| Localization | Estimates own pose | Wheel encoders, IMU, SLAM, markers |
| Planning | Chooses route and motion | Global path planner, local obstacle avoidance |
| Control | Executes commanded motion | Velocity loops in the robot controller |
| Power | Supplies onboard energy | Lithium-ion battery, charging contacts |
Locomotion types and their trade-offs
The choice of locomotion mechanism sets the terrain a mobile robot can handle, its energy efficiency and its mechanical complexity. Wheels dominate on flat floors because rolling contact is efficient, while legs and tracks trade efficiency for the ability to cross rough or stepped ground.
| Locomotion type | Best terrain | Strength | Main limitation |
|---|---|---|---|
| Differential drive | Flat indoor floors | Simple, turns on the spot | Cannot move sideways |
| Omnidirectional wheels | Smooth floors | Moves in any direction | Poor on bumps and debris |
| Ackermann steering | Roads, yards | Stable at speed | Large turning radius |
| Tracked | Mud, rubble, stairs | High traction | Skid steering wastes energy |
| Legged | Uneven, stepped ground | Crosses gaps and steps | Complex control, high power use |
| Aerial rotors | Open air | Ignores ground obstacles | Flight time often under 40 minutes |
A key concept is the holonomic constraint. A differential-drive robot is nonholonomic: it cannot slide sideways, so parking beside a shelf requires a maneuver, much like parallel parking a car. Omnidirectional platforms remove that constraint, which simplifies planning in tight aisles but costs robustness on uneven floors.
Levels of autonomy and navigation
Mobile robots differ most in how they decide where to move. The scale runs from direct human remote control to full autonomy, and the level determines how much perception and computing the robot needs onboard.
| Level | How navigation works | Typical example |
|---|---|---|
| Teleoperated | Human drives through a video link | Bomb disposal robot, telepresence robot |
| Guided path | Follows tape, wire or magnetic markers | Classic AGV in a warehouse |
| Supervised autonomy | Robot drives, human approves or intervenes | Inspection robot in a plant |
| Fully autonomous | Builds or uses a map and plans freely | Autonomous mobile robot, robot vacuum |
Autonomous navigation rests on three questions: where am I, where am I going, and how do I get there. Localization fuses wheel odometry, inertial data and range sensing, because odometry alone drifts with every meter traveled as wheels slip. A map supports a global path, while a local planner reacts to people and obstacles in real time, often recomputing commands 10 to 50 times per second.
How mobile robots relate to neighboring terms
Mobile robot is an umbrella term, and several neighboring terms in robotics fundamentals name subsets of it or overlap with it. Precise use avoids a common confusion between the broad category and its specific forms.
| Term | Relation to mobile robot | Distinguishing feature |
|---|---|---|
| AGV | Subset | Follows fixed guided paths |
| Autonomous mobile robot | Subset | Navigates freely with onboard maps |
| Quadruped robot | Subset | Four-legged locomotion |
| Drone | Subset | Aerial locomotion |
| Service robot | Overlapping category | Defined by task, may be mobile or fixed |
| Industrial robot arm | Contrasting category | Fixed base, bounded workspace |
A mobile manipulator combines both families by mounting a robotic arm on a mobile base. It gains reach across a whole facility but inherits the hardest problems of both: the arm must compensate for base positioning errors of a few centimeters, and the combined system must stay stable while lifting a payload.
Design limits and common pitfalls
Real deployments fail more often on the environment than on the robot itself. The most frequent problems are localization loss, energy limits and unexpected floor or lighting conditions that were absent during testing.
- Battery runtime constrains every design: indoor ground robots often run 8 to 12 hours, while aerial robots typically fly far shorter.
- Reflective, glass and dark surfaces can blind LiDAR and depth cameras, producing phantom or missing obstacles.
- Dynamic environments with people, carts and moved pallets break maps that assumed a static world.
- Floor transitions such as ramps, thresholds and expansion joints cause wheel slip and odometry errors.
- Safety requirements for robots near people demand speed limits, protective fields and emergency stops, which reduce throughput.
A frequent mistake is testing only in a clean laboratory. A robot that navigates perfectly in an empty corridor may stall in a busy aisle, so realistic pilots with real traffic, real lighting and full shift durations are essential before scaling a fleet.
The short version
A mobile robot is a robot that can travel from place to place instead of staying bolted in one spot. A factory arm is like a worker who never leaves the workbench; a mobile robot is like a courier who walks the whole building.
Because it moves, it has to solve problems a fixed robot never faces: knowing where it is, finding a route and avoiding bumping into things. Wheels, legs, tracks and propellers are simply different ways of getting around, each suited to different ground.
Where it makes a difference
A hospital introduces a fleet of four wheeled mobile robots to carry linen, meals and lab samples between departments. Each robot uses LiDAR and a stored map of the corridors, calls elevators through a wireless interface and slows to walking speed near people. Previously, porters spent a large part of each shift pushing carts along the same routes.
During the pilot, the robots stalled at a glass-walled bridge between two buildings, because the LiDAR readings were unreliable there. The team added reflective markers and a reduced-speed zone, and the robots then ran reliably around the clock, docking to charge between deliveries. The porters were reassigned to patient-facing work, which was the outcome the hospital wanted from the project.