Automotive-Specific Dynamic Test Techniques
What the exam tests
Dynamic test techniques used in automotive testing: black-box and white-box methods, fault injection, back-to-back testing, robustness, and timing tests across unit, integration, system, and acceptance levels.
Core rules
- Classical black-box techniques apply: equivalence partitioning, boundary value analysis, decision tables, state transition testing, and use case or scenario testing. Automotive adds environmental, robustness, fault-injection, diagnostic, and end-of-line tests.
- White-box techniques include statement, branch, and MC/DC coverage. ISO 26262 recommends increasingly strong coverage at higher ASILs; MC/DC is associated with ASIL D for certain software levels.
- Fault injection tests error handling and safety mechanisms: corrupted CAN frames, sensor out-of-range values, lost communication, and watchdog timeouts. Back-to-back testing compares model, code, and hardware behavior.
- Timing and real-time behavior are first-class test targets: deadline misses, jitter, task overruns, and startup or shutdown sequences.
Worked example
A coolant temperature sensor reports −40 °C to 150 °C. For boundary value analysis, test −41, −40, 0, 150, and 151 °C. The invalid values check sensor plausibility and fault reaction. For ASIL C software, add branch or MC/DC coverage according to the chosen verification method. Then inject a CAN signal loss to verify the default reaction and diagnostic trouble code. A test set with only nominal values would miss the boundary and fault-handling defects that dominate automotive failures, so the reasoning links each technique to a specific risk.
Traps
- Testing only nominal values and missing boundary and fault behavior.
- Confusing MC/DC with statement coverage.
- Treating back-to-back testing as a substitute for requirements-based tests.
- Forgetting timing, diagnostics, and communication faults.