Parallel Co-Processor Control for Low-Latency Dynamic Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional programmable logic controllers (PLCs) exhibit excessively long latencies and are not suited for dynamic environments requiring rapid and precise responses, such as aircraft control or complex chemical processes, due to their serial execution of instructions and lack of parallel processing capabilities.
Innovation Solution
A control system employing a master processor and multiple asynchronous slave co-processors, where each co-processor evaluates a subset of conditions based on specific criteria like time scale or patterns of change, significantly reducing latency by distributing condition evaluation and allowing reflexive responses to environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional PLCs are used for control tasks, then they can operate in demanding industrial environments with rugged design, but they exhibit excessively long latencies due to serial instruction execution
Solution Approach 1:
The control system is segmented into multiple independent co-processors, each dedicated to evaluating specific conditions or condition subsets. This parallel architecture eliminates the serial execution bottleneck of conventional PLCs while maintaining industrial robustness through distributed condition evaluation.
Solution Approach 2:
The system transitions from single-dimensional serial processing to multi-dimensional parallel processing by introducing multiple co-processors that operate simultaneously on different condition subsets, fundamentally changing the time dimension of control response.
2Device complexity
If conventional PLCs with serial execution are used, then device complexity remains manageable, but response time to environmental changes is excessively slow
Solution Approach 1:
Conditions are segmented and distributed across multiple co-processors, with each co-processor independently evaluating its assigned subset. This segmentation enables parallel condition evaluation, dramatically increasing response speed while keeping individual co-processor complexity manageable.
Solution Approach 2:
Each co-processor performs partial action by evaluating only a specific subset of conditions rather than all conditions sequentially. This partial evaluation approach enables simultaneous condition monitoring across multiple co-processors, achieving excessive speed improvement over serial processing.
3Ease of operation
If conventional PLCs execute instructions sequentially, then programming and operation are straightforward, but productivity in dynamic environments is reduced due to long cycle times
Solution Approach 1:
The control logic is segmented into multiple independent co-processors that evaluate condition subsets in parallel. This segmentation increases control cycle throughput by eliminating sequential execution bottlenecks while maintaining operational simplicity through standardized co-processor interfaces.
Solution Approach 2:
Multiple co-processors are merged into a unified control system that combines their parallel condition evaluation capabilities. This merging achieves high productivity through simultaneous condition monitoring while preserving ease of operation through centralized coordination of the parallel processors.
Data Source
AI summary
A dynamic environment (e.g., an automated industrial process) has multiple conditions in response to which corresponding actions are required, and comprises various equipment, control device(s) to control the equipment, and one or more sensors to generate input signal(s) representing a monitored condition of the environment. A control system for the environment comprises a master processor and one or more co-processors, wherein the master processor configures a given co-processor to evaluate only a first subset of conditions expected to occur in the environment within a specified time period (e.g., less than a response time of the master processor), and to provide first control information representing an action to be taken if a particular condition of the first subset is satisfied. The co-processor receives the input signal(s) representing the monitored condition, processes the input signal(s) so as to determine if the particular condition of the first subset is satisfied, and provides the first control information to the control devices so as to control the equipment. Exemplary applications include dynamic environments in which machine vision techniques and/or equipment are employed.