Service Overview
System Test AI
FPT Automotive's system-test-ai automates system-level validation, HIL/SIL loop simulation, and ISO 26262 functional safety compliance. Integrates AUTOSAR, CANoe, and MATLAB/Simulink into Jenkins/GitLab pipelines and SDV DevOps environments with virtualized SILS boards. Automatically generates test cases, validates edge models through a 3-stage validation pipeline (AI Model → Hardware → Vehicle Loop), and analyzes vision-based inspections to cut execution time by 75%.
75%
Test execution time cut
ISO 26262
Functional safety compliance
18.7×
Inference speedup on hardware
99.7%
Vision inspection accuracy
3 stages
Model/Hardware/Vehicle loop
Capabilities
Key capabilities
End-to-End Test Automation
Comprehensive system-level testing from requirements to validation.
HIL/SIL Support
Full support for Hardware-in-the-Loop and Software-in-the-Loop testing environments.
Traceability Matrix
Complete traceability from requirements to test cases to results.
Automated Test Generation
AI generates system test scenarios based on requirements and specifications.
Performance Optimization
75% reduction in test execution time through intelligent test scheduling.
Technology
Technology stack
| Component | Technology | Purpose |
|---|---|---|
| Test Framework | AUTOSAR, CANoe | Automotive testing |
| Simulation | MATLAB/Simulink, CarMaker | Vehicle simulation |
| Test Generation | LLM, AI models | Test case generation |
| Traceability | Requirements management tools | Test traceability |
| CI/CD | Jenkins, GitLab CI | Automation pipeline |
Use cases
Real-world applications
Documented outcomes from actual deployments.
AI Model Validation Pipeline
Three-stage validation pipeline (AI Model Loop → Hardware Loop → Vehicle Loop) to validate AI models for automotive applications within tight production timelines, ensuring both performance targets and functional safety are met.
Before
No structured pipeline to validate AI models against rigorous automotive performance criteria
After
Both AI models deployed and exceeded production targets on schedule
Vision AI Door Inspection — 2s at 99.7% Accuracy
Automated computer-vision inspection system for door-panel screw verification. System evaluates each unit in real time and returns an immediate pass/fail decision — replacing manual visual checks with an AI-driven test running at production speed.
Before
3-minute manual visual check per door; bottleneck worsened with throughput growth and still missed defects under pressure
After
AI inspection at 99.7% accuracy in 2 seconds; operators only intervene on flagged items
How we work
Implementation approach
Phase 1: Requirements Analysis
- Analyze system requirements and specifications
- Define test scenarios and acceptance criteria
- Plan HIL/SIL test environment setup
Phase 2: Test Generation & Automation
- Generate system test cases from requirements
- Automate test execution in HIL/SIL environments
- Create traceability matrix
Phase 3: Validation & Optimization
- Execute tests and validate results
- Optimize test execution order and scheduling
- Ensure comprehensive requirement coverage
Phase 4: Continuous Integration
- Integrate with CI/CD pipeline
- Automate regression testing
- Monitor test results and quality metrics
Explore more
Related services
Refactoring AI
Automated code review, MISRA-C compliance, and safe API porting for SDV migrations.
Data Analysis & ML
Predictive safety, driver context on Qualcomm, EV diagnostics, and drawing reviews powered by ML.
Unit Test AI
AI unit test generation at 85-90% accuracy, virtualized execution, and MBD check-rule automation.