Construction Machine Terrain Object Detection via Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-propelled construction machines, such as road milling machines and slipform pavers, face challenges in navigating and operating around objects present in the terrain, like manhole covers and curbs, due to limited visibility and accuracy in marking these objects, leading to increased operational complexity and safety margins.
Innovation Solution
Equipping the machines with an image capturing unit and a data processing system that processes images to identify and account for terrain objects, allowing for automatic or manual selection and control adjustments to maintain safe distances and trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual marking of terrain objects is used to guide construction machines, then operators can identify object positions, but the process requires additional work steps, reduces accuracy, and increases operational complexity
Solution Approach 1:
The patent replaces the mechanical manual marking system with an optical sensing system (camera) and automated image processing. The camera captures images of terrain objects, and a control unit automatically processes these images to determine object positions, eliminating the need for manual marking operations while improving position identification accuracy.
Solution Approach 2:
The construction machine performs self-positioning and self-guidance by capturing its own surrounding environment with an integrated camera and processing the images through its control unit. The machine independently identifies terrain objects and uses this information for navigation without requiring external marking operations.
2Ease of operation
If manual marking of terrain objects is used, then object positions can be identified, but visibility is limited and marking is difficult under adverse conditions such as darkness or rain
Solution Approach 1:
The patent replaces manual visual marking with an optical camera system that can capture images under various lighting conditions. The camera, positioned to view the terrain ahead, provides automated object detection that is not limited by human visual capabilities in darkness or adverse weather, improving operational reliability.
3Reliability
If a high margin of safety is selected due to marking inaccuracy, then object avoidance is ensured, but extra work and reduced productivity are required
Solution Approach 1:
The patent replaces manual marking with automated optical detection and image processing that provides precise object position data. This accuracy eliminates the need for large safety margins, allowing the construction machine to operate closer to actual object boundaries while maintaining reliable avoidance, thus improving productivity.
4Ease of operation
If the milling roller is positioned below the cab for terrain alteration, then the working device can access the terrain surface, but the operator cannot see the exact position of terrain objects at the roller level
Solution Approach 1:
The patent introduces a camera as an intermediary sensing device that captures visual information from the terrain ahead and to the sides. The camera is positioned to view areas that are otherwise invisible to the operator, and the control unit processes these images to provide comprehensive object position information for safe operation.
Data Source
AI summary
A self-propelled construction machine has an image capturing unit for capturing an image of the terrain, a display unit on which the captured image is displayed, and a data processing unit. In the displayed image, objects that are relevant to the building project, for example manhole covers, water inlets or curbs, are identified and a spatial data set containing information regarding the position of the identified objects in a reference system which is independent of the movement of the construction machine is determined from the image data of the capturing unit. The construction machine is then controlled in order to carry out translational and/or rotational movements on the terrain and/or to install a structure on the terrain or to alter the terrain on the basis of the spatial data set. As a result, the previously identified objects can be taken into account in the building project.