Trailer Camera Triangulation for Stationary Vehicle Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems cannot accurately detect objects, such as persons on the ground, when a vehicle is at a standstill or moving very slowly, as they rely on vehicle movement to determine depth information and object classification.
Innovation Solution
A method using a trailer camera to capture images from different positions, with the camera's position adjusted using an active actuator system independent of vehicle movement, allowing for accurate determination of object information through triangulation, even when the vehicle is stationary or moving slowly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single camera is used to determine 3D structure by vehicle movement (Structure from Motion), then depth information can be obtained when the vehicle is moving, but object detection fails when the vehicle is at a standstill or moving very slowly
Solution Approach 1:
The system dynamically adapts its operation mode based on vehicle movement state. When the vehicle is moving, it uses Structure from Motion with SfM camera. When the vehicle is stationary or moving slowly, it automatically switches to using the trailer camera with active position adjustment. This dynamic switching resolves the contradiction by ensuring reliable object detection across all vehicle movement states.
Solution Approach 2:
The trailer camera acts as an intermediary system that provides independent depth measurement capability. It is coupled to the towing vehicle but can adjust its position actively using an actuator system, thereby providing a baseline for triangulation independent of vehicle movement. This intermediary system ensures continuous object detection capability regardless of vehicle motion state.
2Measurement precision
If trailer camera position is adjusted using vehicle movement, then depth information can be determined through triangulation, but the system cannot obtain sufficient position changes when the vehicle is stationary or moving slowly
Solution Approach 1:
The trailer camera system uses an active actuator system to dynamically adjust its position independent of vehicle movement. This allows the camera to achieve the necessary position changes for triangulation even when the vehicle is stationary or moving slowly, thereby resolving the contradiction between measurement precision and speed requirements.
Solution Approach 2:
The system separates the camera positioning function from the vehicle movement function. The trailer camera is coupled to the towing vehicle but can be independently positioned using its own actuator system. This segmentation allows the camera to achieve position changes without requiring vehicle movement, enabling accurate spatial observation at any vehicle speed.
3Area of stationary object
If multiple cameras are mounted on both towing vehicle and trailer, then comprehensive environment coverage is achieved, but system complexity increases
Solution Approach 1:
The trailer camera serves multiple functions: it provides depth measurement through triangulation, captures images for object detection, and can be positioned independently to cover different areas. This multi-functionality reduces the need for multiple specialized cameras, thereby achieving comprehensive environment coverage while limiting system complexity increase.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables reliable and accurate spatial observation of the vehicle environment, allowing for object detection and classification, even at low speeds or when stationary, thereby meeting safety requirements for driverless industrial vehicles.
Implementation Method 1
determining object coordinates of the assigned object point from first image coordinates of the at least one first pixel and from second image coordinates of the at least one second pixel by triangulation assuming a base length between the two positions of the trailer camera
Data Source
AI summary
The disclosure relates to a method for determining object information relating to an object in an environment of a multi-part vehicle having at least one towing vehicle and at least one trailer and a control unit and vehicle associated with it, with at least one trailer camera being arranged on the trailer, having at least the following steps: capturing the environment with a trailer camera from a first position and, in dependence thereon, creating a first image having first pixels; changing the position of the trailer camera; capturing the environment with the trailer camera from a second position and creating a second image having second pixels; and, determining object information relating to an object in the captured environment.