LiDAR Signal Encoding for Autonomous Vehicle Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated driving systems face challenges in accurately detecting and responding to objects outside their field of view due to interference from signals, particularly with LiDAR sensors, which can lead to inaccurate path planning and vehicle control.
Innovation Solution
The system employs a controller that differentiates between signals encoded with a first and second scheme, using LiDAR sensors to infer object locations and behavior models, allowing it to adjust vehicle paths based on the presence of objects, even if they are outside the sensor's field of view, by analyzing return signal angles and intensities, and filtering out multipath reflections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR sensors are used to detect objects, then measurement precision is improved, but object-generated harmful factors worsen due to signal interference from other vehicles' LiDAR
Solution Approach 1:
The patent segments the detected objects into two categories: encoded objects (within field of view) and unencoded objects (outside field of view). This segmentation allows the system to apply different processing methods to each category, resolving the contradiction by enabling precise detection of unencoded objects without being misled by signal interference from other vehicles' LiDAR emissions.
Solution Approach 2:
The patent converts the harmful signal interference from other vehicles' LiDAR into a beneficial detection opportunity. By detecting unencoded LiDAR signals that fall outside the vehicle's own field of view, the system transforms what would normally be considered interference or noise into useful information about objects located beyond the sensor's direct viewing angle, thereby improving overall detection accuracy.
2Adaptability or versatility
If the system detects objects outside field of view using unencoded signals, then adaptability is improved, but device complexity worsens due to additional signal processing requirements
Solution Approach 1:
The patent inverts the traditional approach by not filtering out unencoded LiDAR signals as noise, but rather actively detecting and utilizing them. Instead of discarding signals that don't match the expected encoding scheme, the system leverages these unencoded signals to detect objects outside the field of view, thereby expanding detection coverage without requiring additional hardware complexity.
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 enables the system to obtain accurate information about objects proximate to the vehicle, even when they are outside the sensor's field of view, leading to improved path planning and vehicle control decisions, enhancing safety and efficiency.
Implementation Method 1
at least one sensor configured to emit a signal having a first encoding scheme and receive a return signal
Data Source
AI summary
An automotive vehicle includes at least one actuator configured to control vehicle steering, acceleration, or shifting, at least one sensor configured to emit a signal having a first encoding scheme and receive a return signal, and at least one controller in communication with the actuator and the sensor. The controller is configured to control the actuator according to a first mode and a second mode. The controller is further configured to, in response to the sensor receiving a return signal having the first encoding scheme, control the actuator according to the first mode and, in response to the sensor receiving a return signal not having the first encoding scheme, control the actuator according to the second mode.