Valeo Brings Ultrasonic Park Assist to the Mass Market

You are trying to maneuver into a tight parking space and a friendly tone tells you what is ahead and behind the car. The faster the beeps, the closer you are to unseen obstacles. This is the job of the Ultrasonic Park Assist system from Valeo. Where engineers previously spent months painstakingly tweaking and tuning systems on car mock-ups for a handful of top-of-the-line premium automobile models, Valeo has expanded its business into the mass market of average minivans, Sport Utility Vehicles (SUVs), and family cars. Profit margins are much smaller on these vehicles, but the size of the market is huge – representing a growing revenue stream for the company. But this strategy comes with a big challenge: engineers can only spend a fraction of the time designing sensors and mounts for each vehicle model. That is where acoustic simulation comes in, allowing Valeo to develop these innovative acoustic devices in up to only half the time previously required with prototype testing. The Ultrasonic Park Assist system is just one example of Valeo’s strategy to use simulation up-front in future developments of a variety of detection and switching devices, including a window switch currently being designed.

Using acoustic simulation in the development of Valeo’s Ultrasonic Park Assist (UPA) devices is the brainchild of Dr. Richard Rapp, Manager of Simulation at the Switches and Sensing Systems Division. He was hired in 1997 to investigate ways to apply analysis technology at the company, one of the world leaders in automotive traffic sensing and warning systems that today includes ultrasonic parking assistance, blind spot detection, lane departure warning and other driving assistance systems. The company is a tier-one supplier serving nearly every major car company in the world today. Dr. Rapp says he immediately identified the need for performing development work faster to meet the business demands of the lucrative mass market for the park assist system. “It was a natural fit to use simulation to speed up the development of these sensors and mounts, since so much of the work before was performed through extensive testing of physical prototypes,” he notes.

The UPA is elegant in its simplicity, acting like the sonar system in bats and submarines. Excited by a 40-Kilohertz electrical signal, a piezoelectric ceramic membrane mounted in the car’s bumper vibrates at its resonant frequency and emits a sound wave that bounces off objects in the vehicle’s path. The reflected echo is detected by the same membrane, which vibrates and reverses the piezoelectric process – turning sound energy into an electrical signal. The system’s internal circuitry tracks the time taken for the echo and can thus compute the distance from the car to the object.

Dr. Rapp explains that in developing the devices, engineers must establish the amplitude and direction of emitted sound to ensure proper operation. “Too much energy will create spurious secondary echoes that confuse the system,” he says. “Too little energy will not bounce back and would not be detected.” To complicate matters, the emitted sound is affected tremendously by the way the sensor is mounted in the vehicle bumper: the recess depth of the mount, the angle of the funnel-shaped mount opening, the location on the bumper, and the location of all surrounding parts, such as license plates, radiator grill, trailer hitches, trim and overhangs.

Accounting for these many real-world factors takes considerable know-how and time. In the years before simulation was used, engineers would spend considerable time tinkering with physical prototypes of bumper and body mock-ups until they found a configuration that worked. Unfortunately, frequent changes in the car body would force engineers to redesign the system over and over again. And quality problems often did not surface until the final stages of testing, thus requiring quick-fix modifications that did not necessarily make for an optimal design.

To overcome these issues, Dr. Rapp has a strategy of moving into a fully digital development loop through what he calls a “continuous virtual process chain” that would take the design from concept to a final design through virtual prototyping and simulation. “This process allows us to do more engineering up-front in the development cycle to investigate alternatives and refine the design early, rather than fixing problems near the end of the cycle,” explains Dr. Rapp. “Physical prototypes would only be used as a final verification of the design.”

The continuous virtual process chain begins with a modal analysis with ANSYS Finite-Element (FE) software to ensure the sensor membrane’s resonant frequency is near 40 Kilohertz. Next, dynamic calculations are performed in ANSYS to determine the amplitude of vibration. These data are fed into the LMS SYSNOISE acoustic simulation program, which uses Boundary-Element (BE) methods to compute the surface pressure and the resulting sound field produced by the sensor. Through a number of post-processing options, LMS SYSNOISE can provide a variety of outputs, including transient or frequency response functions of sound pressure, acoustic intensities, surface velocities, color contour maps, bar charts, and 2-dimensional directivity polar plots.

According to Dr. Rapp, the information most valuable for the Valeo engineers is a hemispheric surface envelope depicting the shape and intensity of the acoustic field. “LMS SYSNOISE provides an accurate presentation of the sound in three dimensions. This gives our engineers an understanding of the sensor, something which was not possible before,” says Dr. Rapp. “This capability is especially useful in evaluating the effects of the sensor mount and surrounding structure. We need broad radiation in the horizontal plane to cover the whole range behind the car and narrow radiation in the vertical plane to minimize reflection from the ground. Before, working with physical mock-ups involved much testing. Now, our engineers know for sure how the product behaves and can refine the design before prototypes are built.”

Compelling benefits

Acoustic simulation has compelling advantages for Valeo. Dr. Rapp figures the process enables engineers to save up to 50% of the time ordinarily required to develop their product than when using prototype testing alone. “Besides time savings, we also lower costs, improve quality, and obtain greater knowledge of product behavior in order to develop more innovative designs that precisely meet market demand and customer expectation,” he explains. “This allows us to more efficiently develop acoustic sensing systems for the mass market with faster turnaround times, greater economy, and designs that are both optimal and innovative.”

Dr. Rapp also reports that the graphical 3-dimensional output of acoustic simulations enables Valeo to work more closely with its growing list of customers. “Simulation allows us to show our know-how to car companies and impress them with our knowledge about our products’ behavior,” he says. “This capability has tremendous business value in gaining our customers’ trust and confidence in us. Also, from an engineering perspective, simulation gives us quantifiable information to go back to the customer with suggested changes on the configuration of the vehicle bumper and surrounding hardware.”

In addition to the UPA and other traffic environment sensing systems, Valeo’s Switches and Sensing Systems Division also develops a multitude of other devices such as engine sensors, cruise-controls, and instrument-panel switches and modules. 

According to Dr. Rapp, the development of these systems and components will benefit from the continuous virtual process chain, which hinges on a seamless integration of technologies for design and various types of analyses. Valeo has standardized on CATIA V5 for CAD and on ANSYS for detailed FE analysis work. LMS SYSNOISE is heavily used for acoustic simulation. Other technologies will be integrated into the virtual process chain for determining other performance attributes such as system dynamics, durability and vibration.

Dr. Rapp explains that the LMS Virtual.Lab platform is being used to integrate these technologies and provides an ideal link between CATIA, ANSYS, and the different key performance attributes supported in LMS Virtual.Lab. Users can access ANSYS modeling and results as well as make ANSYS an integral part of the LMS Virtual.Lab-supported engineering process. Users also have the ability to automatically set up ANSYS solutions and drive the ANSYS solver through LMS Virtual.Lab.

In applications such as this, LMS Virtual.Lab serves as a common platform in tying together design and analysis programs that otherwise would run separately and produce isolated results – requiring users to spend time reworking models, duplicating mesh representations and copying information from one system to another. LMS Virtual.Lab maintains full associativity across these different applications so that output data from one program can be automatically applied to others – design as well as analysis. This way, changes to the geometry can be propagated in an automated manner through the entire analysis sequence, alternative design variations can be efficiently evaluated, and engineers can more effectively balance trade-offs between different attributes.

As part of this integration strategy, Valeo has also benchmarked LMS Virtual.Lab Motion, multibody dynamic software for analysis of motion and loads of mechanical systems. LMS Virtual.Lab Motion was selected because of its ability to work seamlessly with CATIA V5 and ANSYS, its integration into the Virtual.Lab platform, the capability to readily handle flexible body structures, its hierarchical tree structure, ease-of-use, and the excellent technical support from LMS.

In the multibody simulation benchmark, a child safety protection window switch was analyzed to reduce excessive switching forces required in moving the spring-loaded wheel of an actuator assembly into a ramped quadrant where it locks into position. With previous trial-and-error experimentation, engineers would modify either the spring constant or the quadrant shape, one at a time. This procedure yielded workable designs in the past, but not necessarily the best configuration. With multibody simulation in combination with the LMS Virtual.Lab Optimization module, all parameters can be optimized at the same time, resulting in an optimal design that satisfies all design criteria. In this case, LMS Virtual.Lab was able to reduce switching forces by 30% with a refined design that without the help of simulation would probably not have been achieved.

“As we integrate these simulation technologies into the virtual process chain, Valeo demonstrates a leadership role in innovatively leveraging these tools. Utilizing the latest technology is not an objective by itself, however, but rather an effective way of strengthening its commanding position in the competitive automotive supply chain.”