Dual Recognizer Sensor Data Transmission for Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transmission techniques for self-driving cars face challenges in efficiently collecting training data due to bandwidth limitations and power consumption, leading to potential delays in AI updates and reduced object detection accuracy.
Innovation Solution
An information processing apparatus that employs a first and second recognizer to determine transmission necessity and priority based on recognition results, prioritizing data transmission with the second recognizer's more accurate but resource-intensive processes, especially when computational resources are available, to efficiently collect training data for AI updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the second recognizer with higher accuracy is used for data transmission determination, then object detection accuracy is improved, but power consumption and computational resource usage increase
Solution Approach 1:
The system dynamically switches between the first recognizer (lower accuracy, lower power consumption) and the second recognizer (higher accuracy, higher power consumption) based on driving conditions and transmission necessity. This dynamic adjustment allows the system to optimize the balance between detection accuracy and power consumption in real-time.
Solution Approach 2:
The system changes the operational parameters by selecting different recognizers with different capability conditions. The controller determines whether to use the second recognizer based on transmission necessity and priority, effectively changing the system's detection capability parameters to match the current operational requirements.
2Quantity of substance
If all sensor information is transmitted to the server apparatus, then training data completeness is improved, but transmission time and network bandwidth consumption increase
Solution Approach 1:
The system extracts and transmits only the necessary sensor information that requires AI model updates, rather than transmitting all sensor data. By using the two recognizers to identify discrepancies and determine transmission priority, the system extracts only the critical subset of data needed for effective training data collection.
Solution Approach 2:
The system transmits sensor information partially - only when determined necessary by the comparison between first and second recognition results. This partial action approach avoids the excessive transmission of all data while ensuring that essential training data is transmitted.
3Measurement precision
If the second recognizer is always used, then AI model accuracy is improved, but device complexity and computational load increase
Solution Approach 1:
The system employs a dynamic architecture where the second recognizer is activated only when needed for determining transmission necessity and priority. The controller dynamically decides when to engage the more complex second recognizer based on the first recognition result and transmission requirements, reducing overall computational complexity.
Solution Approach 2:
The recognition function is segmented into two levels: the first recognizer handles routine detection with lower computational requirements, while the second recognizer is invoked only when transmission determination is needed. This segmentation reduces the overall computational load while maintaining high accuracy when necessary.
Data Source
AI summary
An information processing apparatus is configured to cause a first recognizer to execute a first recognition process that takes sensor information as input, and a second recognizer to execute a second recognition process that takes the sensor information as input, the second recognizer being configured to operate under different capability conditions from the first recognizer; determine one of a transmission necessity and a transmission priority of the sensor information depending on a difference between a first recognition result of the first recognition process and a second recognition result of the second recognition process; and transmit the sensor information to a server apparatus according to the determined one of the transmission necessity and the transmission priority.