Vehicular Control Simulation with Deterministic Task Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern vehicular systems face challenges in achieving deterministic simulations due to nondeterminism arising from system time and multiple processes, leading to inconsistent simulation outputs and difficulty in debugging and reproducing defects.
Innovation Solution
A method for simulating vehicular control systems using discrete diagrams with execution policies that enforce determinism by controlling task execution and data flow, ensuring consistent simulation results through policies like strong causality, weak causality, and snapshot isolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If nondeterministic execution policies are used to allow flexible task scheduling, then system adaptability is improved, but simulation consistency deteriorates
Solution Approach 1:
The system changes the execution policy parameter from nondeterministic to deterministic based on the simulation mode. In deterministic simulation mode, the system enforces strict execution ordering where tasks are executed only when all their input data is available, eliminating timing-related nondeterminism while maintaining task scheduling flexibility through the discrete event simulation framework.
2Ease of operation
If system time is used for task scheduling, then ease of operation is improved, but simulation determinism deteriorates
Solution Approach 1:
The system extracts and removes the source of nondeterminism (system time) from the simulation environment. Instead of using system time for task scheduling, the simulator uses virtual simulation time that is independent of the host system's clock, thereby eliminating timing variations while maintaining simple task scheduling through the discrete event simulation mechanism.
3Productivity
If multiple processes are used to simulate different vehicle functions, then productivity is improved, but simulation consistency deteriorates
Solution Approach 1:
The system segments the simulation into independent discrete events that can be processed in deterministic order. Each vehicle function is represented as separate tasks in the discrete diagram, and the simulator processes these tasks event-by-event in a predetermined sequence, ensuring that parallel processing does not introduce nondeterminism.
4Ease of operation
If execution order is not controlled, then ease of operation is improved, but debugging capability deteriorates
Solution Approach 1:
The system performs preliminary action by pre-defining the execution order of tasks through the discrete diagram structure before simulation begins. The discrete diagram explicitly specifies the sequence and dependencies of tasks, allowing defects to be reproduced consistently by following the predetermined execution path without requiring complex runtime control mechanisms.
Data Source
AI summary
A method for testing a vehicular control system includes receiving, at a simulator, a discrete diagram identifying a plurality of tasks to be simulated and connections between the plurality of tasks, and receiving, at the simulator, a requested determinism level selected at a user device from a plurality of determinism levels. An optimal execution policy based on the simulator, the discrete diagram and the requested determinism level is selected from a plurality of execution policies. The plurality of tasks are simulated, by the simulator, using the selected optimal execution policy to simulate operation of the vehicular control system on a vehicle.