# How the iRobot Roomba Automatically Cleans Floors Without Falling Down Stairs > A 2002 patent describing an autonomous floor-cleaning robot that uses sensors to navigate rooms, avoid cliffs like stairs, and sweep debris into a removable bin. - **Patent:** US 6883201 - **Original title:** Autonomous floor-cleaning robot - **Owner:** iRobot Corp - **Granted:** 2005 - **Status:** Public domain (expired) - **Times cited:** 215 - **Field:** consumer_electronics, mechanical, ai_ml ## What it does This patent describes an autonomous robot that cleans floors by combining a motive system for movement, a vacuum, and a primary brush assembly. It features a side brush that reaches beyond the robot's main body to pull dirt into the cleaning path. The robot uses a sensor system, specifically a cliff detector, to identify stairs or ledges and prevent itself from falling. When a cliff is detected, the control system automatically adjusts the robot's speed and direction to escape the area and resume cleaning. ## What it does NOT cover - Does not cover non-robotic manual vacuum cleaners or push-brooms. - Does not cover cleaning systems that rely on external navigation markers or beacons installed in a room. - Does not cover robots that use cameras or visual mapping (SLAM) to navigate, as this patent focuses on tactile and cliff sensors. - Does not cover robots that utilize liquid-based mopping systems as their primary cleaning mechanism. ## The clever bit The innovation lies in the integration of a side brush that extends beyond the chassis, combined with a control system that treats 'cliffs' as a distinct sensor event requiring a specific escape maneuver rather than just a standard obstacle. ## Real-world examples 1. Original iRobot Roomba models 2. Roomba 400 series 3. Early autonomous vacuum cleaners ## Why it matters This patent is a foundational document for the modern consumer robotics industry. It established the core behavioral logic—such as the 'bounce' mode and cliff avoidance—that allowed the original iRobot Roomba to become a household staple. It effectively defined the 'bump-and-go' navigation style that dominated the market for over a decade. ## Frequently asked questions ### What does How the iRobot Roomba Automatically Cleans Floors Without Falling Down Stairs cover? A 2002 patent describing an autonomous floor-cleaning robot that uses sensors to navigate rooms, avoid cliffs like stairs, and sweep debris into a removable bin. ### Who owns patent US 6883201? iRobot Corp owns this patent, granted in 2005. ### When does this patent expire? This patent has expired and is now in the public domain — anyone can use the invention freely. ### What is patent US 6883201 cited by? This patent has been cited by 215 later patents that build on its ideas. ### What problem does this patent solve? This patent is a foundational document for the modern consumer robotics industry. It established the core behavioral logic—such as the 'bounce' mode and cliff avoidance—that allowed the original iRobot Roomba to become a household staple. It effectively defined the 'bump-and-go' navigation style that dominated the market for over a decade. ### What does this patent NOT cover? Does not cover non-robotic manual vacuum cleaners or push-brooms. **Full plain-English explainer:** https://patentbrief.org/patent/us/6883201/roomba-autonomous-vacuum **Original patent:** https://patents.google.com/patent/US6883201 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How James Spangler Invented the First Portable Electric Vacuum Cleaner](https://patentbrief.org/patent/us/889823/vacuum-cleaner-spangler) — A 1908 patent for a portable suction-based cleaning device that combined a rotating brush with a fan to lift dust into a bag. - [How AI Learns to Control Game Characters Based on Their Surroundings](https://patentbrief.org/patent/us/10607134/artificially-intelligent-systems-devices-and-methods-for-learning-andor-using-an-avatars-circumstances-for-autonomous-avatar-operation) — A system that allows digital characters to automatically perform actions by matching their current environment to previously learned experiences stored in a database. - [How Self-Balancing Vehicles Warn Users Before They Tip Over](https://patentbrief.org/patent/us/6302230/segway-personal-transporter) — A safety system for self-balancing vehicles that monitors how much 'room' the machine has left to accelerate before it loses its ability to stay upright. - [How AI Agents Learn to Pick the Best Future Actions](https://patentbrief.org/patent/us/10282665/action-selection-with-a-reward-estimator-applied-to-machine-learning) — A method for an AI agent to predict which actions will yield the highest rewards by analyzing past experiences and refining its decision-making model. - [How George Devol Invented the First Industrial Robot Arm](https://patentbrief.org/patent/us/2988237/unimate-industrial-robot-devol) — The 1954 patent for the Unimate, the first digitally controlled robotic arm that could be programmed to move objects in a factory.