Drivable Area Detection with Camera-Radar Probability Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle drivable area detection methods in autonomous driving technologies are not accurate enough, leading to incomplete recognition of driving environments and potential collisions with obstacles.
Innovation Solution
A method that integrates a camera apparatus and a radar to process image data and radar echo signals using a neural network, fusing the obtained probability distributions to represent the drivable area of a vehicle as a probability, thereby enhancing the accuracy of obstacle detection and avoiding blind spots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single sensor (camera or radar) is used for obstacle detection, then the device complexity is reduced, but the measurement precision and detection accuracy deteriorate due to blind spots and limited detection ranges
Solution Approach 1:
The patent combines a camera apparatus and a radar into an integrated sensor system. The camera captures image data while the radar detects obstacles using electromagnetic waves, and their results are fused to produce a comprehensive probability distribution map of obstacles. This merging approach compensates for the blind spots and limitations of individual sensors, achieving higher detection accuracy without excessive complexity increase.
2Ease of operation
If conventional single-sensor methods are used, then the system is simpler to operate, but the reliability of drivable area recognition deteriorates due to incomplete environment perception
Solution Approach 1:
The patent introduces a neural network as an intermediary that automatically fuses data from multiple sensors and generates a probability distribution map of obstacles. This intermediary component handles the complex data processing and integration tasks, allowing the system to maintain high reliability through multi-sensor fusion while keeping the operation simple for the user. The neural network transforms raw sensor data into actionable probability information automatically.
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 method provides a more comprehensive understanding of the drivable area around a vehicle, improving navigation planning flexibility and enhancing the reliability of autonomous driving systems by accurately recognizing both drivable and non-drivable areas.
Implementation Method 1
obtaining a second probability distribution of the obstacle based on a time of flight and an echo width of a radar echo signal
Implementation Method 2
a radar are integrated, obstacle distribution information obtained by the two sensors is fused
Data Source
AI summary
This application discloses a vehicle drivable area detection method, an autonomous driving assistance system, and an autonomous driving vehicle. The method includes: processing, by using a neural network, image data obtained by a camera apparatus, to obtain a first probability distribution of an obstacle; obtaining a second probability distribution of the obstacle based on a time of flight and an echo width of a radar echo signal; and obtaining, based on the first probability distribution of the obstacle and the second probability distribution of the obstacle, a drivable area of a vehicle represented by a probability, where the probability is a probability that the vehicle cannot drive through the area. The autonomous driving assistance system includes a camera apparatus, at least one radar, and a processor. The system is configured with the technical solutions that can implement the method. The autonomous driving vehicle includes the foregoing autonomous driving assistance system.