Camera-Based Boom Control Using Neural Network Pose Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for controlling articulating boom arms in machinery like feller bunchers and excavators rely on sensors for determining the pose of the boom, which can be unreliable or fail, leading to inaccuracies in positioning and movement.
Innovation Solution
The integration of a camera system that uses a neural network to determine the pose of the articulating boom arm based on image data, supplementing or replacing sensor data to ensure accurate positioning and movement, even in the event of sensor failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensors are used to determine the pose of the boom arm, then the positioning accuracy can be maintained, but the system reliability deteriorates due to sensor failure or unreliability
Solution Approach 1:
The system implements a neural network-based vision system as a pre-established backup mechanism that can take over pose determination when sensors fail. The neural network is trained offline and ready to provide positioning accuracy as a cushion against sensor failure, thereby maintaining both reliability and measurement precision.
Solution Approach 2:
The patent introduces a camera-based vision system as an intermediary mechanism between the boom arm and the control system. This vision system processes images to determine boom pose, serving as a mediator that can replace faulty sensors and maintain positioning accuracy without direct dependence on unreliable sensor components.
2Reliability
If a camera system with neural network is integrated to determine boom pose, then the system reliability improves through redundancy, but the device complexity increases
Solution Approach 1:
The neural network is trained offline in advance using labeled image data before deployment. This preliminary training action prepares the system to directly infer boom pose from images during operation, reducing online computational complexity and making the integrated system more manageable despite the added hardware.
Solution Approach 2:
The system creates a digital model of the boom arm pose by training a neural network to recognize and interpret visual features in camera images. This copying approach translates physical boom positions into image-based representations, allowing the system to determine pose through image analysis rather than direct sensor measurement, thereby improving reliability while managing complexity through software-based solutions.
3Device complexity
If sensor data is used as the primary mechanism for determining boom pose, then the device complexity is kept simple, but the reliability deteriorates when sensors fail or are missing
Solution Approach 1:
The system dynamically changes the input parameters for pose determination based on sensor availability. When sensors are functional, it uses sensor data (simpler mode). When sensors fail or are missing, it switches to using camera image data processed by the neural network (reliable mode). This parameter switching resolves the contradiction by adapting to conditions rather than requiring a permanently complex system.
Data Source
AI summary
Systems and methods are described for determining an actual pose of an articulating boom arm using an artificial intelligence mechanism (e.g., a neural network) trained to determine the actual pose of the articulating boom arm based on captured image data. In some implementations, an electronic processor is configured to control movement of the articulating boom arm based at least in part on pose information determined by applying the image-based neural network. In some implementations, an electronic processor is configured to train the neural network by using, as training data, captured image data and output signals from sensors indicative of measured positions of the components of the articulating boom arm.