Online ML Control for Autonomous Earth Moving Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The operation of earth moving vehicles is costly and challenging due to the need for manual operators and the complexity of adapting to changing soil conditions and varied vehicle models, which complicates the development of effective autonomous control systems.
Innovation Solution
An autonomous or semi-autonomous earth moving system that uses a combination of sensors and machine learning models to control and navigate earth moving vehicles, selecting optimal tool paths and control signals based on real-time data from the vehicle and environment, allowing for adaptive operation across different soil conditions and vehicle types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual operators control earth moving vehicles, then operational flexibility and adaptability to changing conditions are maintained, but operational costs increase and productivity decreases
Solution Approach 1:
The earth moving vehicle is equipped with sensors and machine learning models that enable it to autonomously sense changing soil conditions, determine optimal operating parameters, and adjust its own control signals without human intervention. This self-service capability allows the vehicle to maintain adaptability while improving productivity through continuous operation and elimination of human limitations
Solution Approach 2:
The patent replaces manual mechanical control with an autonomous control system comprising sensors, processors, and actuators. The system uses machine learning models to process sensor data and generate control signals, substituting human operators with an automated intelligent system that can continuously adapt to changing conditions while maintaining high productivity
2Productivity
If autonomous control systems are implemented, then productivity and operational efficiency improve, but system complexity increases due to varying soil conditions and vehicle models
Solution Approach 1:
The patent employs a universal machine learning model architecture that can handle multiple earth moving vehicle types and varying soil conditions through a single unified system. The system uses sensor data from multiple sources and processes it through trained models to generate appropriate control signals for different vehicle configurations, eliminating the need for separate control systems for each vehicle type
Solution Approach 2:
The system dynamically adjusts control parameters based on real-time sensor data and machine learning model predictions. By changing operational parameters such as tool path, speed, and actuation signals based on detected soil conditions and vehicle state, the system maintains simplicity in control architecture while achieving adaptability through parameter optimization
3Device complexity
If traditional control methods are used for earth moving vehicles, then system simplicity is maintained, but adaptability to changing soil conditions and vehicle models deteriorates
Solution Approach 1:
The patent replaces traditional mechanical and manual control methods with an intelligent control system based on machine learning models. These models are trained on data from multiple vehicle types and soil conditions, enabling the system to automatically adapt to different vehicle configurations and environmental conditions without requiring complex manual programming or reconfiguration
Data Source
AI summary
An autonomous earth moving system can determine a desired state for a portion of the EMV including at least one control surface. Then the EMV selects a set of control signals for moving the portion of the EMV from the current state to the desired state using a machine learning model trained to generate control signals for moving the portion of the EMV to the desired state based on the current state. After the EMV executes the selected set of control signals, the system measures an updated state of the portion of the EMV. In some cases, this updated state of the EMV is used to iteratively update the machine learning model using an online learning process.