Domestic Robot Visual Docking Calibration Using Base Station Markers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Domestic robotic systems, such as vacuum cleaners and lawnmowers, face challenges in accurately determining their position and orientation within their working areas, which is crucial for efficient navigation and autonomous charging, as existing methods like triangulation and GPS can be inaccurate and require complex calculations.

Innovation Solution

A domestic robotic system equipped with an image obtaining device and sensors, where a processor detects predetermined patterns associated with markers, allowing for accurate position and orientation estimation, and performs calibration processes to refine navigation systems, enabling precise navigation to charging stations or other destinations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If triangulation or GPS receivers are used to determine robot position and orientation, then navigation capability is provided, but measurement precision and reliability are insufficient for accurate docking

Engineering Contradiction:
Improveposition and orientation determination accuracyVSAvoidnavigation system reliability for docking
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces markers as intermediary objects placed on the base station and predetermined locations in the working area. These markers serve as mediators between the robot's navigation system and the environment, providing high-precision reference points for position and orientation determination through image processing, thereby achieving accurate docking without relying on less precise triangulation or GPS systems

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates a simplified geometric model (copy) of the environment by detecting marker positions and orientations in images. This geometric model is then used for navigation and docking operations, replacing complex real-world environmental analysis with a manageable computational representation that enables precise position determination

Inventive Principle:
Principle #26Copying

2Measurement precision

If mechanical setups are used to guide robot to correct position and orientation, then docking accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvedocking accuracyVSAvoidmechanical guidance system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces mechanical guidance systems with an optical-computational system. Instead of using mechanical structures to physically guide the robot to the correct position, the system uses image obtaining devices to capture marker images, processes these images to determine position and orientation, and controls the movement system accordingly. This substitution eliminates complex mechanical guidance mechanisms while achieving equivalent or superior docking accuracy

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The robot performs self-navigation and self-docking by autonomously detecting markers, calculating its position and orientation relative to the base station, and adjusting its movement accordingly. The system eliminates the need for external mechanical guidance structures by enabling the robot to determine its own position and guide itself to the correct docking position through image processing and computational geometry

Inventive Principle:
Principle #25Self-service

3Measurement precision

If visual sensing systems are used for marker detection, then position estimation is achieved, but noise affects measurement precision

Engineering Contradiction:
Improveposition estimation accuracyVSAvoidnoise in visual sensing
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary geometric constraints and mathematical models to the marker detection process. By pre-defining the geometric relationships between markers and their expected positions in the image plane, the system can filter out noise and outliers. The geometric model provides a framework for validating detected marker positions and rejecting measurements that do not conform to the expected geometric patterns, thereby improving position estimation accuracy despite visual sensing noise

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3168705B1Domestic robotic system
Publication Date: 2021.09.15 F ROBOTICS ACQUISITIONS
  • EP3168705B1 patent drawingFigure 1~2A
  • EP3168705B1 patent drawingFigure 2B~2C
  • EP3168705B1 patent drawingFigure 3A

AI summary

A domestic robotic system, for example for mowing the lawn, including a robot, which includes: a movement system having wheels or the like for moving the robot over a surface; an image obtaining device, such as a camera, for obtaining images of the exterior environment of the robot; and at least one processor in electronic communication with the movement system and the image obtaining device. The at least one processor is programmed to: detect a predetermined pattern within at least one of the images, with this predetermined pattern being associated with a marker provided on a base station; respond to the detection of the predetermined pattern by determining, by a first process, an estimate of the robot's position and/or orientation, this estimate of the robot's position and orientation being relative to the base station, the at least one processor and the image obtaining device thereby forming part of a first positioning system for the robot; determine, by a second process, an alternative estimate of the robot's position and/or orientation, the at least one processor thereby forming part of a second positioning system for the robot; and perform at least one calibration of the second positioning system using the first positioning system.