Mobile Robot Navigation Using Image-Guided Waypoint Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robotic systems lack intuitive and efficient navigation methods, particularly in dynamic environments, making them difficult for users to operate effectively.

Innovation Solution

A method and system for navigating mobile robots using image data from multiple cameras, a graphical user interface (GUI) on an operator device, and data processing hardware to determine a pointing vector and transmit waypoint commands, enabling the robot to navigate to a target location while avoiding obstacles and adjusting for terrain estimates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If traditional robotic navigation systems are used, then the robot can move autonomously, but the ease of operation is poor and users cannot intuitively guide the robot

Engineering Contradiction:
Improveease of operationVSAvoiddevice complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The system creates a virtual copy of the physical environment by capturing images with the robot's camera and displaying them on a graphical user interface. Users interact with this virtual representation by selecting pixels on the displayed image, which corresponds to physical locations in the real environment. This copying approach allows intuitive control without requiring users to understand complex robotic navigation commands or coordinates.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The graphical user interface acts as an intermediary between the user and the robot's navigation system. Instead of directly controlling the robot through complex commands, users interact with the GUI by selecting pixels on the displayed image. The system then translates these pixel selections into navigation commands, mediating the interaction and simplifying the control process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If the robot navigates to distant target locations, then the navigation capability is improved, but the system cannot account for terrain variations and obstacles

Engineering Contradiction:
ImproveadaptabilityVSAvoidmeasurement precision
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system performs preliminary terrain estimation and ground plane detection before executing navigation to distant targets. By analyzing the captured images and estimating the ground plane in advance, the system can predict terrain variations and adjust the navigation path accordingly, ensuring accurate arrival at the target location even over long distances.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors the robot's position and the terrain characteristics during navigation. By comparing the actual terrain with the estimated ground plane, the system can detect deviations and adjust the navigation path in real-time, maintaining measurement precision throughout the journey to distant targets.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4010771B1Navigating a mobile robot
Publication Date: 2026.03.04 BOSTON DYNAMICS INC
  • EP4010771B1 patent drawingFigure 1
  • EP4010771B1 patent drawingFigure 2
  • EP4010771B1 patent drawingFigure 3A

AI summary

A method (1700) for controlling a robot (300) includes receiving image data (342) from at least one image sensor (344). The image data corresponds to an environment (301) about the robot. The method also includes executing a graphical user interface (221) configured to display a scene (222) of the environment based on the image data and receive an input indication indicating selection of a pixel location (224) within the scene. The method also includes determining a pointing vector based on the selection of the pixel location. The pointing vector represents a direction of travel for navigating the robot in the environment. The method also includes transmitting a waypoint command to the robot. The waypoint command when received by the robot causes the robot to navigate to a target location. The target location is based on an intersection between the pointing vector and a terrain estimate of the robot.