Sensor Fusion Pipeline for Over-Drivable Object Height Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor fusion systems for autonomous or semi-autonomous vehicles face challenges in accurately estimating the height and over-drivability of stationary objects, such as debris, in a vehicle's path, as they rely solely on camera or radar data, which can lead to inaccurate range and angular resolution, resulting in potential collisions.
Innovation Solution
A two-stage sensor fusion pipeline that combines time-series radar data with camera images using a histogram tracker to project radar range detections onto pixels, enabling accurate pixel-based width and height estimation of stationary objects, thereby facilitating safe over-drivability decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sensor fusion systems rely solely on camera or radar data for object detection, then device complexity is reduced, but measurement precision of object dimensions deteriorates
Solution Approach 1:
The patent combines radar and camera data fusion to estimate object dimensions. Radar provides accurate range measurements while camera provides angular information, and merging these data sources enables precise height and width estimation of stationary objects without requiring complex hardware correlations
2Measurement precision
If expensive hardware is used for sensor data correlation and fusion, then measurement precision of object position improves, but device complexity and cost increase
Solution Approach 1:
The system uses the inherent capabilities of existing radar and camera sensors to provide complementary information for dimension estimation. Rather than requiring expensive specialized hardware, the solution leverages the self-service properties of standard sensor fusion components already present in autonomous vehicles
3Measurement precision
If radar range detections are applied to image-based dimensions, then measurement precision of object height and width improves, but processing time increases
Solution Approach 1:
The patent performs preliminary radar range detections and camera image capture simultaneously, then applies the range data to the image-based dimensions in a coordinated manner. This preliminary action approach allows the system to prepare dimension estimation data in advance, reducing processing delays when collision avoidance decisions are needed
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
This approach enhances the accuracy of object dimension estimation, allowing vehicles to quickly and safely navigate around or over obstacles by combining the strengths of radar range detection with camera-based angular determinations, improving vehicle autonomy and safety.
Implementation Method 1
receiving a first electromagnetic wave within a first frequency band reflected from an object within a path of a vehicle
Data Source
AI summary
This document describes techniques, apparatuses, and systems for sensor fusion for object-avoidance detection, including stationary-object height estimation. A sensor fusion system may include a two-stage pipeline. In the first stage, time-series radar data passes through a detection model to produce radar range detections. In the second stage, based on the radar range detections and camera detections, an estimation model detects an over-drivable condition associated with stationary objects in a travel path of a vehicle. By projecting radar range detections onto pixels of an image, a histogram tracker can be used to discern pixel-based dimensions of stationary objects and track them across frames. With depth information, a highly accurate pixel-based width and height estimation can be made, which after applying over-drivability thresholds to these estimations, a vehicle can quickly and safely make over-drivability decisions about objects in a road.