Aging Indices for Sensor Fusion in Time-Controlled Architectures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed time-controlled real-time computer architectures, integrating pre-processing results from various sensors with different processing times into sensor fusion is challenging due to temporal inconsistencies, as the duration of calculations depends on the sensor data, leading to incomplete results within predetermined frames.
Innovation Solution
Assigning aging indices to pre-processing results based on their availability, weighting them accordingly, and implementing strategies to handle delayed results through frame duration adjustments and data correction, ensuring timely or consistent sensor fusion based on application requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pre-processing algorithms are used with running times depending on input data, then processing flexibility and adaptability are improved, but temporal consistency and synchronization in the time-controlled architecture deteriorate
Solution Approach 1:
The system performs preliminary actions by assigning aging indices to pre-processing results based on their completion time relative to the frame start. This allows the system to pre-categorize results by their temporal freshness (AI=0 for completed within frame, AI=1 for next frame, AI=n for nth frame after), enabling subsequent sensor fusion to account for temporal variations without disrupting the time-controlled architecture's synchronization.
Solution Approach 2:
The invention changes the parameter of temporal consistency by introducing aging indices that quantify the time delay of pre-processing results. By transforming the temporal inconsistency into a measurable parameter (AI value), the system can adaptively weight and fuse sensor data from different time points, resolving the contradiction between flexible processing times and temporal consistency requirements.
2Stability of the object's composition
If sensor fusion waits for all pre-processing results to be available, then temporal consistency is improved, but response time and productivity deteriorate
Solution Approach 1:
The system applies partial action by allowing sensor fusion to proceed with available pre-processing results rather than waiting for all results. Results with AI=0 (completed within the frame) are fused immediately, while results with AI>0 are either fused with appropriate weighting or trigger frame duration increases. This partial execution maintains temporal consistency for available data while improving response time.
Solution Approach 2:
The invention introduces dynamics by making the frame duration adjustable based on pre-processing completion status. The monitor process dynamically increases frame duration when a threshold percentage of results have AI≥1, allowing the system to adapt between waiting for all results (high consistency) and proceeding with available results (fast response) based on real-time processing conditions.
3Reliability
If the frame duration is increased to allow completion of all pre-processing, then completeness and reliability are improved, but system response speed and productivity deteriorate
Solution Approach 1:
The system dynamically adjusts frame duration based on the completion status of pre-processing tasks. The monitor process evaluates the percentage of results with AI≥1 and increases frame duration only when necessary (when threshold P is exceeded). This dynamic adjustment ensures reliability by allowing sufficient processing time when needed, while maintaining fast response speeds when processing completes within the standard frame duration.
Solution Approach 2:
The invention changes the frame duration parameter adaptively rather than using a fixed value. By monitoring the distribution of aging indices and adjusting the frame duration parameter accordingly, the system optimizes the trade-off between completeness (reliability) and response speed, ensuring that frame duration is extended only when processing requirements demand it.
Data Source
AI summary
The invention relates to a method for integration of calculations having a variable running time into a distributed time-controlled real-time computer architecture, which real-time computer architecture consists of a plurality of computing nodes, wherein a global time having known precision is available to the computing nodes, wherein at least part of the computing nodes is equipped with sensor systems, in particular different sensor systems for observation of the environment, and wherein the computing nodes exchange messages via a communication system, wherein at the start of each cyclical frame Fi having the duration d the computing nodes collect raw input data by means of a sensor system, wherein the start times of the frame Fi are deduced from the progress of the global time, and wherein the pre-processing of the raw input data is carried out by means of algorithms, the running times of which depend upon the input data, and wherein the value of the ageing index AI=0 is assigned to a pre-processing result which is produced within the frame Fi at the start of which the input data were collected, and wherein the value of the ageing index AI=1 is assigned to a pre-processing result which is produced within the frame F following the frame in which the input data were collected, and wherein the value AI=n is assigned to a pre-processing result which is produced in the nth frame after the data collection, and wherein the ageing indices of the pre-processing results are taken into consideration in the computing nodes which carry out the fusion of the pre-processing results of the sensor system.