Control Decision Logic From Dynamic Models With Lower Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex engineered systems, such as self-driving cars and planes, face challenges in generating effective control decision logic due to the complexity of dynamic environments, leading to high computational costs and latency issues.
Innovation Solution
The automatic generation of control decision logic using a dynamic physical model, refined through discretization, pipeline staging, and pruning functions, enables efficient control decision-making by reducing precision requirements and optimizing computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional control logic is used for complex engineered systems, then control accuracy can be maintained, but computational cost and processing time increase significantly
Solution Approach 1:
The patent segments the continuous physical state space into discrete regions using spatial partitioning techniques. By dividing the complex control problem into smaller discrete segments, the system can process control decisions more efficiently while maintaining accuracy within each segment. This segmentation allows parallel processing and reduces the computational burden of evaluating continuous physical models in real-time.
Solution Approach 2:
The patent transforms physical parameters into discrete symbolic representations, changing the parameter space from continuous to discrete. This parameter transformation enables the use of efficient discrete logic operations instead of computationally intensive continuous calculations, reducing processing requirements while preserving the essential control relationships through the physical model.
2Measurement precision
If high-precision control logic is implemented, then system control accuracy improves, but response latency increases
Solution Approach 1:
The patent performs preliminary discretization and spatial partitioning of the control space during system initialization or offline processing. By pre-processing the physical model and creating discrete representations in advance, the system eliminates the need for complex real-time calculations during actual control operations, significantly reducing response latency while maintaining control accuracy through the pre-computed discrete logic.
3Reliability
If detailed physical models are used for control, then control accuracy improves, but system complexity increases
Solution Approach 1:
The patent creates simplified discrete copies or representations of the continuous physical model. Instead of directly implementing complex continuous physical equations in control logic, the system uses discrete symbolic copies that capture the essential behavior and relationships of the physical model. These discrete copies maintain the predictive capabilities needed for reliable control while being significantly simpler to implement and process.
Data Source
AI summary
Possible input value combinations of a prediction of an engineered system are iterated over, comprising, for a possible input value combination: selecting an action to perform on the engineered system for the possible input value combination, comprising: performing a plurality of predictions of the engineered system scored by evaluating an objective function associated with the engineered system and using the possible input value combination and a corresponding plurality of actions. The action is selected from the corresponding plurality of actions, the selection being based at least in part on scores of the plurality of predictions. A rule specifying a corresponding set of one or more rule conditions that is met when the possible input value combination is matched and a corresponding action associated with the rule as a selected action is generated. The generated set of rules to be stored or further processed is output.