Working to Achieve Platooning Safety Through Verification and Validation - Peloton Technology
Customer Drivers Trained to Platoon on Highway in One Day
June 7, 2019
How We Got Here: Laying the Groundwork for the Future of Trucking
August 16, 2019
Show all

Working to Achieve Platooning Safety Through Verification and Validation

By: Kieran Donnelly, Quality Assurance Manager
Kieran Donnelly, Quality Assurance Manager

Peloton Technology has made it abundantly clear (for instance, by putting safety first, explaining the regulations we follow, in our safety report, and through rigorous testing) that safety in our truck platooning technology is a non-negotiable priority.

But how do we prove that we follow through on our good intentions?

One important way is through our verification and validation (V&V) efforts, which are part of our larger quality assurance (QA) program.

Peloton takes a pragmatic, practical approach to V&V. We start with a detailed V&V plan, which we present to our technology partners for their sign-off. We then conduct rigorous testing, including regulatory-required and non-mandated tests that verify and validate each aspect of the system.

Extensive Testing That Exceeds Industry Requirements

The starting point for safety testing in our industry is International Standards Organization (ISO) 26262, titled “Road vehicles—functional safety,” which defines functional safety for automotive equipment, including all automotive electronic and electrical safety-related systems.

In our V&V process, we test all the safety mechanism of the Peloton truck platooning technology. To accomplish this we developed Software-in-Loop (SIL) and Hardware-in-Loop (HIL) infrastructure to run automated tests and simulations. We then conduct vehicle integration and system testing on a closed track to further verify and validate the safety and non-safety platooning functions. The data gathered from vehicle testing is analyzed by our development and test engineers using automated tools to identify issues not readily observable by either driver or tester.

Each of these steps (SIL, HIL, and Vehicle) in the testing process constitutes a test gate for which we have established detailed entry, exit, and success criteria. If a test fails or any defect or anomaly is uncovered, then we evaluate, fix, and retest until all test gates have passed.

Following a few months of pre-testing, it takes about 6 weeks to complete the V&V process over tens of thousands of track and highway miles. During this phase, we are validating the platooning system and paying very close attention to the user experience, evaluating how professional truck drivers feel about using the Peloton technology. Does the braking feel too rough? Are the auditory alerts useful? Do they interfere with other sounds in the truck cab? Is the graphical user interface of the Peloton screen convenient and easy to use? Are the buttons to start and stop platooning located in the right place in the cab?

Drivers and test engineers provide feedback on all aspects of the user experience, including during daylight, at night, on flat and hilly roads, and in various weather conditions. We incorporate the feedback to make sure the Peloton system works for all kinds of drivers in the environments in which they will platoon.

When we have completed all the tests outlined in the V&V plan, we assemble the results and submit a report for review both internally at Peloton and to our partners, which include some of the top-tier players in the automotive industry with extensive experience in safety testing. So far, our partners have all approved Peloton’s rigorous V&V testing, with compliments.

Boring = Successful

Peloton’s goal in all our quality assurance work is to do testing on our time, not on our customers’ time.

What we test, and how we test it, is determined through a tight feedback loop among Peloton’s product development, QA, and user experience (UX) teams. We encourage everyone involved to file suggestions as feedback, and we take a data-driven approach to incorporating that feedback into the product.

The reality, though, is that successful V&V requires a lot of time developing and executing automated and manual tests while analyzing and documenting the results from thousands of monotonous highway miles. At least ideally they’re monotonous. Safety testing is not the time to seek a colorful experience.

If we do our jobs right, the process will be the opposite of exciting. It will be downright boring. And boring, to us, is the best sign of V&V success.