Remote Robot Control With On-the-Fly Learning in Changing Terrain

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Fully automatic control of devices such as robots in sandy environments like beaches or farmland is challenging due to short-term and long-term environmental changes, leading to inaccuracies and inefficiencies in tasks like weed removal and garbage collection.

Innovation Solution

A method of real-time human-operated control of remote devices using a machine learning algorithm trained on-the-fly with user input, combining human and algorithmic inputs to enhance accuracy and precision, allowing for continuous adaptation to environmental changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If fully automatic control is used, then device operation is automated, but accuracy and reliability deteriorate due to environmental changes

Engineering Contradiction:
Improveautomation levelVSAvoidcontrol reliability
Core Design Contradiction:
Extent of automationVSReliability

Solution Approach 1:

The patent introduces an algorithm as an intermediary between the automatic control system and the environment. The algorithm processes sensor data and generates control signals, acting as a mediator that adapts to environmental changes while maintaining automation. This resolves the contradiction by enabling automatic operation with improved reliability through the algorithm's ability to handle environmental variability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements dynamic control where the algorithm continuously adapts its parameters and decisions based on real-time environmental conditions. The control system transitions from static pre-programmed commands to dynamic responses that adjust to changing sand conditions, weather, and task requirements, thereby maintaining reliability while fully automated.

Inventive Principle:
Principle #15Dynamics

2Speed

If fully automatic control is used, then operational speed increases, but measurement precision deteriorates in changing environments

Engineering Contradiction:
Improveoperational speedVSAvoidenvironmental detection precision
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent implements feedback loops where sensor measurements of environmental conditions are continuously fed back to the algorithm, which adjusts control signals accordingly. This feedback mechanism maintains measurement precision by constantly recalibrating based on actual environmental conditions, enabling fast operation without sacrificing detection accuracy.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The algorithm performs preliminary analysis of environmental data and predicts optimal control actions before executing them. This anticipatory approach allows the system to maintain high operational speed by pre-computing responses to expected environmental changes, while still achieving precise measurement through continuous sensor monitoring.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If human operator control is used, then accuracy and reliability improve, but productivity decreases due to manual operation

Engineering Contradiction:
Improvecontrol accuracyVSAvoidoperational efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent enables the system to serve itself by having the algorithm autonomously process sensor data and generate control signals without continuous human intervention. The algorithm learns from operational data and improves its performance over time, maintaining high accuracy while dramatically increasing productivity through autonomous operation.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical human operator with an algorithmic control system that processes information and generates commands electronically. This substitution eliminates the productivity limitations of manual operation while maintaining or exceeding human-level accuracy through sophisticated sensor processing and adaptive control algorithms.

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

4Measurement precision

If more training data is collected for algorithm training, then algorithm accuracy improves, but loss of time increases due to data collection and processing

Engineering Contradiction:
Improvealgorithm accuracyVSAvoidtraining time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements continuous learning where the algorithm is trained on new data throughout operational periods rather than requiring separate training phases. Sensor data collected during normal operation is continuously processed to refine algorithm parameters, maintaining high accuracy without stopping productivity. This continuous action eliminates the time loss associated with batch training while progressively improving algorithm performance.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP4441572B1A method of real-time controlling a remote device, and training a learning algorithm
Publication Date: 2025.11.05 ODD BOT BV
  • EP4441572B1 patent drawingFigure 1A~1B
  • EP4441572B1 patent drawingFigure 2

AI summary

A method is provided of real-time controlling a remote device to perform a task, the method comprising steps of: for controlling the remote device to perform a task, obtaining graphical data, such as image frames forming a video, of surroundings of the remote device, such as an area of farmland or beach, sending the graphical data to a remote operation device, obtaining user input data from an operator, which user input data is indicative of a location of interest in the graphical data, generating a control signal for controlling the remote device to perform a task based on the user input data, and using the control signal for controlling the remote device to perform the task at the location of interest. The user input data is further used as training data for training a machine learning algorithm, which algorithm is arranged for generating at least part of a control signal for controlling the remote device; and/or providing a suggested location of interest to the operator.