Robotically Steered Sensor Fusion for Long-Range Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems struggle to accurately detect and identify objects at a distance due to resolution limitations of onboard sensors, which can hinder efficient and safe vehicle control.
Innovation Solution
Employing multiple imaging and ranging sensors with varying fields of view and resolutions to capture and enhance object recognition, including a first sensor with a wider field of view and a second sensor with a narrower, higher resolution to improve object identification, particularly using steerable mechanisms for precise imaging of objects of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a single sensor with wide field of view is used, then the coverage area is improved, but the measurement precision of distant objects deteriorates
Solution Approach 1:
The system divides the sensing function into multiple sensors: a first sensor (camera) with wide field of view for detecting objects in the distance, and a second sensor (LIDAR) with narrower field of view for obtaining precise depth information. This segmentation allows each sensor to specialize in its strength, resolving the contradiction between coverage area and measurement precision.
Solution Approach 2:
The system transitions from two-dimensional image data from the camera to three-dimensional spatial information by integrating LIDAR depth data. This dimensional enhancement allows the system to maintain wide coverage while achieving precise measurement of distant objects through the addition of depth dimension.
2Measurement precision
If a single sensor with high resolution is used, then the measurement precision is improved, but the field of view deteriorates
Solution Approach 1:
The system segments the sensing tasks between two sensors: the camera provides wide field of view for initial object detection, while the LIDAR provides high-resolution depth measurement for identified objects of interest. This segmentation allows high precision without sacrificing overall field of view coverage.
Solution Approach 2:
The camera performs preliminary detection to identify objects of interest within the wide field of view, then the LIDAR is activated to obtain precise depth information for those specific objects. This preliminary action approach allows high-resolution sensing to be applied only where needed, maintaining overall wide coverage.
3Measurement precision
If multiple sensors are used to improve object recognition, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The system uses a camera for both wide-area surveillance and as a trigger for LIDAR activation. The camera serves multiple functions: initial object detection, object tracking, and triggering depth sensing. This multi-functionality reduces the need for separate dedicated sensors for each function, thereby reducing overall system complexity.
Solution Approach 2:
The camera performs preliminary object detection and filtering, identifying only objects of interest that require further analysis. This preliminary action reduces the number of objects that need high-resolution depth sensing, thereby reducing the overall computational and hardware complexity of the multi-sensor system.
Data Source
AI summary
Disclosed are methods, systems, and non-transitory computer readable media that control an autonomous vehicle via at least two sensors. One aspect includes capturing an image of a scene ahead of the vehicle with a first sensor, identifying an object in the scene at a confidence level based on the image, determining the confidence level of the identifying is below a threshold, in response to the confidence level being below the threshold, directing a second sensor having a field of view smaller than the first sensor to generate a second image including a location of the identified object, further identifying the object in the scene based on the second image, controlling the vehicle based on the further identification of the object.