Loader Augmented Control for Learned Repetitive Task Cycles

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Power machines, such as compact loaders, face challenges in efficiently performing repetitive tasks due to the need for continuous operator intervention, which can lead to inefficiencies and increased time requirements.

Innovation Solution

The implementation of an augmented control system that allows power machines to learn and autonomously perform repetitive tasks by initiating a learning mode, setting a home position, and recording operations, enabling the machine to automatically repeat the task cycle as specified.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If continuous operator intervention is used to control power machines for repetitive tasks, then the machine can perform tasks with flexibility and adaptability, but the efficiency and productivity are reduced due to time consumption and operator workload

Engineering Contradiction:
ImproveefficiencyVSAvoidoperator intervention
Core Design Contradiction:
ProductivityVSExtent of automation

Solution Approach 1:

The power machine performs repetitive tasks autonomously after the operator demonstrates the task once. The controller records the operator's control inputs and machine responses during a demonstration cycle, then automatically executes the recorded sequence without continuous operator intervention, allowing the machine to serve itself in performing repetitive operations

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The controller creates a digital copy of the operator's demonstrated task cycle by recording control lever positions, machine responses, and operational parameters. This recorded copy is then replayed automatically to execute the same task sequence repeatedly without requiring the operator to physically replicate the actions each time

Inventive Principle:
Principle #26Copying

2Productivity

If augmented control system with learning mode is implemented, then the machine can autonomously perform repetitive tasks improving efficiency, but the device complexity increases

Engineering Contradiction:
Improveoperational efficiencyVSAvoidcontrol system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The existing controller is enhanced to perform multiple functions: normal real-time control, demonstration mode recording, and automated replay execution. By making the controller universal and multi-functional rather than adding separate dedicated systems, the patent improves productivity while minimizing the increase in device complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Loss of time

If the power machine autonomously performs tasks by learning from demonstration, then time is saved in repetitive operations, but the initial setup and learning process requires additional time

Engineering Contradiction:
Improvetime for repetitive tasksVSAvoidlearning mode duration
Core Design Contradiction:
Loss of timeVSDuration of action of moving object

Solution Approach 1:

The operator performs a one-time demonstration of the task cycle during which the controller records all control inputs and machine responses. This preliminary action of recording during demonstration prepares the system for rapid automated execution, sacrificing initial setup time to gain significant time savings during subsequent repetitive task executions

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11372405B2Augmented loader controls
Publication Date: 2022.06.28 DOOSAN BOBCAT NORTH AMERICA INC
  • US11372405B2 patent drawing
  • US11372405B2 patent drawing
  • US11372405B2 patent drawing

AI summary

Disclosed embodiments include power machines or loaders, and systems used on loaders, configured to augment the control of the loader to accomplish repetitive tasks. Also disclosed are methods of learning a task for augmented control of a loader, and methods of controlling a loader to perform a learned task to provide augmented control of the loader.