Practice makes better - Learning effects of driving with a multi-stage collision warning

Area of Science:

• Human-computer interaction
• Traffic safety research
• Automotive engineering

Background:

• Advanced driver assistance systems (ADAS) are crucial for enhancing traffic safety.
• Existing research often focuses on single-use case effectiveness, neglecting diverse urban traffic scenarios.
• Adaptive, multi-stage warning systems require examination across various use cases and over time.

Purpose of the Study:

• To investigate the effectiveness of a multi-stage collision warning system in a driving simulator.
• To analyze driver adaptation and benefit from practice with the adaptive warning system.
• To evaluate driver behavior and subjective ratings across different warning stages and trial repetitions.

Main Methods:

• A driving simulator experiment involving 24 participants.
• Four trials: baseline (no assistance), three assisted trials including a learning phase and a transfer drive.
• Measurement of driving behavior (brake reaction time, brake force) and subjective ratings in response to multi-stage warnings (W1, W2) across varied safety-critical scenarios.

Main Results:

• Drivers exhibited faster brake reactions and utilized appropriate braking force across both warning stages (W1, W2) compared to the baseline.
• Performance improvements were observed successively over the assisted trials, indicating a learning effect.
• Drivers successfully transferred learned behaviors to new scenarios, demonstrating adaptability and benefit from practice.

Conclusions:

• The two-stage collision warning system is effective in mitigating accidents by promoting earlier and stronger decelerations.
• Driver practice with ADAS, even in simulated environments, is essential for maximizing safety benefits.
• The adaptive, multi-stage warning system is well-accepted by users and can be recommended for practical implementation.