Autonomous Guidance System Small Object Detection via Radar Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous guidance systems face challenges in accurately detecting and navigating around small objects, such as curbs and lamp-posts, which are typically ignored by radar systems and require critical camera alignment for image-based systems to determine distance reliably.
Innovation Solution
An autonomous guidance system integrating a camera module and a radar module with a controller that classifies objects as small based on the magnitude of the radar reflection signal, allowing for the generation and updating of detailed maps to include small objects, enhancing navigation and safety by avoiding unexpected small objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If radar systems are used for autonomous guidance, then detection range and speed are improved, but small objects such as curbs and lamp-posts cannot be detected
Solution Approach 1:
The patent combines radar and camera systems into an integrated autonomous guidance system. The radar provides fast detection and ranging information, while the camera provides detailed visual information for small object detection. The controller fuses data from both sensors to achieve both speed and precision in object detection.
Solution Approach 2:
The camera acts as an intermediary that complements the radar system. While radar provides initial detection and ranging, the camera captures detailed images that enable the system to identify and locate small objects that radar cannot detect individually. The controller uses the camera as a mediator to enhance the radar's limited small object detection capability.
2Measurement precision
If image based systems are used for autonomous guidance, then small object detection capability is improved, but critical camera alignment is required to reliably determine distance
Solution Approach 1:
The radar system serves as an intermediary that provides ranging information to compensate for the camera's inability to reliably determine distance without critical alignment. The radar's distance measurement capability supplements the camera's visual information, allowing the system to detect small objects and determine their distances without requiring precise camera alignment.
Solution Approach 2:
The patent merges radar and camera systems where the radar provides robust distance measurement that does not depend on camera alignment. This combination allows the system to maintain small object detection capability while eliminating the need for critical camera alignment, thereby reducing device complexity.
3Reliability
If radar systems ignore small objects to reduce false alarms, then system reliability is improved, but navigation safety around small objects deteriorates
Solution Approach 1:
The patent combines radar and camera systems where the camera detects small objects that radar would otherwise ignore. The controller fuses data from both sensors, using the camera's ability to detect small objects like curbs and lamp-posts while maintaining the radar's reliability in overall system operation. This integration ensures navigation safety around small objects without compromising system reliability.
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
The system effectively determines the location of small objects and integrates radar and image data for improved autonomous vehicle guidance, enabling safer navigation and avoiding obstacles that might otherwise cause vehicle instability or damage.
Implementation Method 1
a radar module to output a reflection signal indicative of a reflected signal reflected by the object
Data Source
AI summary
An autonomous guidance system that operates a vehicle in an autonomous mode includes a camera module, a radar module, and a controller. The camera module outputs an image signal indicative of an image of an object in an area about a vehicle. The radar module outputs a reflection signal indicative of a reflected signal reflected by the object. The controller determines an object-location of the object on a map of the area based on a vehicle-location of the vehicle on the map, the image signal, and the reflection signal. The controller classifies the object as small when a magnitude of the reflection signal associated with the object is less than a signal-threshold.