Sensor Data Compression Using Priority Masks for LiDAR Lag Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in processing and transmitting large amounts of sensor data, leading to lag time and increased risk of incorrect data decoding, especially when using LiDAR sensors, due to the inefficiencies in existing data compression techniques.
Innovation Solution
A data compression system that generates and applies masks to sensor data using artificial intelligence and user input, prioritizing the most significant information for transmission, thereby reducing processing and transmission loads and improving data precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all sensor data is compressed and transmitted, then data completeness is maintained, but processing time and transmission time increase significantly
Solution Approach 1:
The patent segments sensor data into multiple priority levels (first priority, second priority, third priority) based on importance. High-priority data is transmitted and processed immediately, while lower-priority data is processed subsequently or discarded, thereby reducing overall processing time while maintaining essential information completeness.
Solution Approach 2:
The patent extracts and transmits only the most critical data elements (high-priority data) from the complete sensor dataset. By selecting and transmitting only essential information rather than all data, processing and transmission time is reduced while maintaining the completeness of critical information needed for autonomous driving decisions.
2Loss of information
If all sensor data is compressed and transmitted, then data completeness is maintained, but transmission data volume increases
Solution Approach 1:
The patent divides sensor data into priority segments and transmits only high-priority data through the compression system. This segmentation approach reduces the volume of data requiring transmission while ensuring that complete and accurate transmission of critical information is maintained.
Solution Approach 2:
The patent extracts only the essential high-priority data elements from the complete sensor dataset for transmission. By filtering out non-essential lower-priority data, the transmission data volume is reduced while maintaining the completeness of transmitted information.
3Productivity
If traditional compression techniques are used on all data, then processing load is reduced, but decoding accuracy decreases
Solution Approach 1:
The patent applies different processing qualities to different data segments based on their priority. High-priority data receives higher quality processing and transmission to maintain decoding accuracy, while lower-priority data receives reduced processing. This local differentiation of quality maintains accuracy for critical data while reducing overall processing load.
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 enhances the speed and accuracy of data transmission in autonomous driving by prioritizing critical data, reducing lag time, and minimizing errors in decoding, allowing for more responsive and precise object detection.
Implementation Method 1
A LiDAR sensor emits beams of light at predetermined intervals. The emitted beams contact objects and are reflected back to the sensor.
Implementation Method 2
Based on a time between emission and detection, the LiDAR sensor is able to determine the distance between the object and the sensor.
Data Source
AI summary
A system includes a non-transitory computer readable medium configured to store instructions thereon; and a processor connected to the non-transitory computer readable medium. The processor is configured to execute the instructions for generating a mask based on received data from a sensor, wherein the mask includes a plurality of importance values, and each region of the received data is designated a corresponding importance value of the plurality of importance values. The processor is configured to execute the instructions for encoding the received data based on the mask; and transmitting the encoded data to a decoder for defining reconstructed data. The processor is configured to execute the instructions for computing a loss based on the reconstructed data, the received data and the mask. The processor is configured to execute the instructions for providing training to an encoder for encoding the received data based on the computed loss.