LIDAR Blind Spot Detection Using Virtual Sensor Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-based navigation systems for self-driving cars face challenges in accurately detecting objects, particularly in blind spots, due to sensor limitations and occlusions, which can lead to delayed corrective actions and potential collisions.
Innovation Solution
A method and system utilizing a Machine Learning Algorithm (MLA) to process LIDAR point cloud data, analyzing LIDAR points and grid representation data to identify blind spots, and employing remedial actions such as adjusting vehicle movement or acquiring additional sensor data to correct detection issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensors (cameras, LIDARs, radars) are used to detect objects, then detection coverage is improved, but blind spots still exist due to occlusions and sensor limitations
Solution Approach 1:
The patent introduces a virtual sensor as an intermediary element that mediates between physical sensors and the object detection system. The virtual sensor is generated through machine learning algorithms that predict sensor readings from LIDAR point cloud data, effectively filling blind spots where physical sensors cannot detect objects due to occlusions or line-of-sight limitations
Solution Approach 2:
The patent creates a virtual copy of sensor data by generating synthetic sensor readings that mimic physical sensor outputs. The machine learning model learns to generate virtual sensor measurements from LIDAR point clouds, producing a copy of what sensor data would look like if the physical sensor could directly observe the target, thereby compensating for blind spots
2Speed
If the computer system detects objects faster, then corrective actions can be taken sooner, but detection accuracy may be compromised in blind spots
Solution Approach 1:
The patent performs preliminary actions by generating virtual sensor data and detecting potential objects in blind spots before the vehicle reaches critical distances. The machine learning model continuously predicts sensor readings and identifies potential hazards in advance, allowing the control system to prepare corrective actions sooner while maintaining accuracy through the virtual sensing mechanism
3Reliability
If the vehicle slows down or stops to avoid collisions, then safety is improved, but productivity and travel efficiency decrease
Solution Approach 1:
The system performs preliminary detection and assessment of potential hazards using virtual sensors, allowing the vehicle to maintain normal speed when no threats are detected. By continuously monitoring blind spots and identifying risks in advance, the system only triggers speed reductions or stops when necessary, thereby maintaining productivity while improving safety
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
Enhances the detection of objects around self-driving cars by identifying blind spots and enabling timely corrective actions, improving safety and reducing the risk of collisions.
Implementation Method 1
a LIDAR having a plurality of lasers and configured for capturing LIDAR point cloud data
Implementation Method 2
capturing LIDAR point cloud data having a plurality of LIDAR points
Data Source
AI summary
A method of and system for processing Light Detection and Ranging (LIDAR) point cloud data. The method is executable by an electronic device, communicatively coupled to a LIDAR installed on a vehicle, the LIDAR having a plurality of lasers for capturing LIDAR point cloud data. The method includes receiving a first LIDAR point cloud data captured by the LIDAR; executing a Machine Learning Algorithm (MLA) for: analyzing a first plurality of LIDAR points of the first point cloud data in relation to a response pattern of the plurality of lasers; retrieving a grid representation data of a surrounding area of the vehicle; determining if the first plurality of LIDAR points is associated with a blind spot, the blind spot preventing a detection algorithm of the electronic device to detect presence of at least one object surrounding the vehicle conditioned on the at least one object is present.