Camera-Based Work Machine Distance Ranging for Uneven Terrain
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Off-road work machines, particularly those operating in non-flat terrains, face challenges in accurately predicting distances to objects due to measurement errors, leading to potential damage to the machines and surrounding objects.
Innovation Solution
A work machine management system that determines a range of potential distances to objects based on image analysis, considering the height and uncertainty associated with the object type, and adjusts operations accordingly to maintain safe distances, triggering safety actions when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distance prediction is based on flat surface assumption using camera features, then the calculation is simple and fast, but the prediction accuracy deteriorates in non-flat terrains
Solution Approach 1:
The system changes the parameter representation from a single predicted distance value to a probability distribution over multiple possible distance values. This allows the system to maintain computational efficiency while capturing the uncertainty and variability inherent in non-flat terrain conditions, effectively resolving the contradiction between calculation speed and prediction accuracy.
Solution Approach 2:
Instead of attempting to perfectly predict the single true distance, the system performs partial prediction by generating a set of plausible distance values with associated probabilities. This excessive action of predicting multiple outcomes rather than one provides more information for safe operation, particularly by identifying the shortest possible distance to take conservative safety actions.
2Device complexity
If a single predicted distance is used for safety decisions, then the control logic is simple, but safety may be compromised due to measurement errors
Solution Approach 1:
The system applies beforehand cushioning by using the shortest distance in the predicted distribution as a conservative estimate for safety decisions. This pre-empts potential measurement errors by assuming the worst-case scenario (closest object distance), ensuring that safety actions are triggered even if the true distance is shorter than the most likely prediction. This resolves the contradiction by maintaining simple control logic while significantly improving safety assurance.
3Productivity
If the work machine operates autonomously without distance verification, then operation efficiency is high, but collision risk increases
Solution Approach 1:
The system implements feedback by continuously monitoring the predicted distance distribution and comparing it against safety thresholds. The autonomous operation is maintained high-level, but with layered feedback mechanisms: the full probability distribution provides rich information for intelligent decision-making, while the shortest distance provides a simple feedback signal for immediate safety actions. This resolves the contradiction by enabling autonomous operation with intelligent risk assessment rather than simple binary control.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A system and a method are disclosed for determining a range of potential distances between a work machine and an object. The system receives an image captured by a camera on the work machine and identifies an object in the image. The system determines an angle between the camera and the object, a height associated with the object, and an uncertainty associated with the height. Based on the angle, the height, and the uncertainty, the system determines a range of potential distances between the work machine and the object. The shortest distance in the range is compared to a threshold distance for safe operation of the work machine. When the shortest distance in the range is less than the threshold distance, the system causes the work machine to perform a safety action.