Vehicle Inference Timing Using Longer-Interval Time Series Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In automated driving, the accuracy of periodically executed inference operations using time series data is low when the interval of acquiring time series data is equal to the execution cycle of the inference operation.
Innovation Solution
An information processing apparatus that executes an inference operation by inputting time series data to a neural network, where the interval of acquiring constituting data of the time series data is longer than the execution cycle, allowing for improved recognition of changes in environmental data and enhanced accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the interval of acquiring time series data is equal to the execution cycle of the inference operation, then the system operates efficiently with synchronized timing, but the accuracy of the inference operation is low
Solution Approach 1:
The system performs preliminary data acquisition at a longer interval before the inference operation executes. By acquiring time series data at intervals longer than the execution cycle, the system prepares higher-quality input data that captures more significant environmental changes, thereby improving inference accuracy without requiring more frequent execution cycles
Solution Approach 2:
The system changes the time interval parameter of data acquisition from being equal to the execution cycle to being longer than the execution cycle. This parameter change allows the system to capture more meaningful temporal variations in environmental data, improving the quality of input for the neural network while maintaining the original execution frequency
2Measurement precision
If the interval of acquiring time series data is longer than the execution cycle, then the accuracy of inference operation is improved by magnifying changes in environmental data, but the synchronization between data acquisition and operation execution becomes less frequent
Solution Approach 1:
The system performs preliminary data acquisition at a longer interval before the inference operation executes. By acquiring time series data at intervals longer than the execution cycle, the system prepares higher-quality input data that captures more significant environmental changes, thereby improving inference accuracy without requiring more frequent execution cycles
Solution Approach 2:
The system maintains periodic execution of the inference operation at the original execution cycle frequency, while the data acquisition operates at a different periodic interval (longer than the execution cycle). This allows the system to preserve its productivity and response frequency while using less frequent, higher-quality data inputs
Data Source
AI summary
An information processing apparatus includes a processing unit configured to execute an inference operation in an execution cycle. The inference operation is executed by inputting input data including time series data to a neural network. An interval of acquiring constituting data of the time series data to be input in a single time of the inference operation is longer than the execution cycle.