Autoliv Lower Body Restraint System

Team: Taylor Torgerson

Project Description

The Autoliv company wants to continue to improve airbag systems for use in vehicles. To continue their innovation, Autoliv requires more accurate testing technology for improvements. They have created a testing setup, with a sled and a human model. This setup mimics the front compartment of a car during a crash to test potential chest airbags. The setup currently only includes a chest airbag, which differs from reality because modern cars have several other airbags. The absence of additional airbags causes test inaccuracies.

Autoliv wants to improve the accuracy of a linear sled crash test for chest airbags by including a knee airbag. Cost constraints prevent Autoliv from using a real knee airbag in every crash test. The team was tasked with creating a system that limits lower body motion in the same way a real knee airbag would during the crash test.

Linear Sled Crash Test Setup — Linear sled crash test setup with dummy and restraint system

Comparison of targets, thresholds, predicted performance, and actual performance for each system requirement. Asterisk (*) indicates an estimate. Double asterisk (**) indicates the system was not completed in time to measure actual performance.
Requirement / Constraint / Goal	Target	Threshold	Predicted Performance	Actual Performance
Test-to-Test Pelvis Travel Distance	N/A	5 mm	5 mm*	3 mm
Available Restrained Pelvis Travel Distance	N/A	50–250 mm	50–250 mm*	100–200 mm
Available Free Pelvis Travel Distance	N/A	0–100 mm	0–100 mm*	100 mm
System Tunability	10 mm increments	50 mm increments	10 mm increments*	10 mm
System Compatibility	Fit any PAB LAT Fixtures	Fit 2 PAB LAT Fixtures	Fit any PAB LAT Fixtures	**
Subsystem Individual Weight	50 lb	100 lb	30 lb	~40 lb
Reusability	300 tests	100 tests	300 tests*	300 tests*
Initial Leg Angle Range	N/A	25–70°	25–70°	25–70°
Tibia Restraint Angle	90°	90° ± 10°	90° ± 10°	90° ± 10°
Test Dummy Compatibility	Fit with 5th, 50th, 95th percentile test dummies	Fit with 5th, 50th percentile test dummies	Fit with 5th, 50th, 95th percentile test dummies	Fit with 5th, 50th, 95th percentile test dummies
Consumable Price Per Test	$20	$200	$30	$20

* Indicates an estimate.
** System was not completed in time to measure actual performance.

Design Process

The restraint system is composed of four subsystems: the damper, pivot, knee, and rope. As the linear sled test is run, a “crash” occurs and the dummy lurches forward. As the dummy moves forward, the knee subsystem is pulled forward. The rope gets loaded in tension and transfers the force back towards the damper. The pivot system directs the loading, and the damper dissipates the impact energy by deforming a metal sample.

Knee Subsystem

The purpose of this subsystem is to create a point of contact between the dummy and the rest of the system. A metal tube is strapped onto the dummy’s legs, and the rope connects through holes on either end of the tube.

Pivot Subsystem

The purpose of this subsystem is to change the direction of the rope, allowing it to be pulled at 90 degrees to the dummy’s tibia and level with the damper.

Damper Subsystem

The purpose of this subsystem is to deform a metal sample as the center slider is pulled forward. This deformation creates the restraint that decelerates the dummy’s lower body.

Rope Subsystem

The purpose of this subsystem is to transfer the energy from the dummy’s movement to the damper.

Testing

Once the system design had been completed, the team selected a variety of metal bar samples with varying heights and thicknesses. Three rounds of impact testing were performed to evaluate the sample performance under different speeds.

Figure 1 — Knee Airbag Displacement vs. Acceleration — **Figure 1 — Knee Airbags.** Displacement vs. acceleration graph depicting the behavior of real knee airbags. X-axis: displacement (mm), approximately 600–750 mm. Y-axis: acceleration (g), approximately −40 to 0.

Figure 2 — Sample Deformation Displacement vs. Acceleration — **Figure 2 — Samples.** Displacement vs. acceleration graph depicting the behavior of the samples as they were deformed within the damper. X-axis: displacement (mm), approximately −400 to 0. Y-axis: acceleration (g), approximately −35 to 0.

The sample deformation and restraint distance test data was compiled, and correlations were used to create a test matrix. The matrix will inform technicians of the required sample for a linear sled test given a desired speed, dummy size, and restraint distance.

Test Matrix — Sample Selection by Speed, Dummy Size, and Restraint Distance — Test matrix: maps desired speed, dummy size, and restraint distance to the required metal sample for a linear sled test.

Conclusion

The design met most of the requirements, though additional testing and analysis are needed to confirm compliance with the remaining criteria. This project showed that consistent communication, early feasibility checks, and expert input are critical for a successful design. The team learned that iteration and testing are necessary, and that interpreting data often requires outside guidance. The team’s design showed that restraining the dummy from behind is a viable option. Future work should focus on expanded testing with varied sample designs, improving system geometry for better fit and performance, exploring more effective cable materials, and redesigning the knee subsystem for greater durability and compatibility.

Special Thanks to Aaron Treglown at Autoliv.