Your behavior appears to be a little unusual. Please verify that you are not a bot.


Taking GNSS receiver testing to new heights

August 24, 2020  - By
Photo: NovAtel

Photo: NovAtel

An applications engineer and his sky-jumping bud don wingsuits to test a NovAtel GNSS receiver integrated with an Epson IMU.

In September 2019, a specialized team assembled at an airstrip outside of Edmonton, Alberta, Canada. Their mission: Put the Hexagon | NovAtel PwrPak7D-E2 enclosed receiver through tricky test procedures that involved jumping out of an airplane at 10,000 feet.

Taking the NovAtel SPAN receiver to the skies was the brainchild of Andrew Levson, who is both a NovAtel engineer and a skydiving aficionado. He proposed using a wingsuit to test the receiver’s positioning accuracy.

The first wingsuit dive took place in 2011, with NovAtel’s OEM615 receiver and ALIGN heading technology.

This time, the engineers aimed to test both NovAtel’s GNSS receiver featuring SPAN tightly coupled GNSS+INS functionality and its new companion, the Epson G370 inertial measurement unit (IMU). Both are packed in the PwrPak7D-E2 to provide uninterrupted positioning even in GNSS-denied environments.

Wingsuit jumpers Andrew Levson (right) and Blair Egan suit up for the NovAtel tests. (Photo: NovAtel)

Wingsuit jumpers Andrew Levson (right) and Blair Egan suit up for the NovAtel tests. (Photo: NovAtel)

“We chose to revive the project, given that equipment has evolved with more comprehensive capabilities,” said Patrick Casiano, manager of Product Management and Applied Technology, NovAtel. “Between 2011 and 2019, we could significantly reduce the payload while increasing value in the data.” In 2011, NovAtel was only able to monitor Levson’s heading. In 2019, the team captured heading, azimuth, pitch and roll measurements.

“We wanted to prove that our equipment can work in a high-dynamic environment, which isn’t necessarily ideal conditions for collecting positioning data,” explained Kiera Fulton, associate product manager, Enclosures and Post-Processing Software, NovAtel. “By proving our products work in a less-than-ideal environment, we exemplify how robust our solutions are.”

Photo: NovAtel

Photo: NovAtel

Test Preparation

For the 2019 test, the team chose to gather attitude data. The team also asked Levson to perform specific skydiving maneuvers to rigorously test the positioning solution. “Rather than performing just a simple flight to the ground, we wanted to challenge the solution to reveal more,” Casiano said.

The test was not easy to implement. A lot of behind-the-scenes planning and preparation went into the project. Plus, unforeseen factors made the test more challenging, Fulton said, such as logistics and weather.

“The skydivers require specific weather conditions in order to jump safely,” Fulton said. “Considering how quickly the weather can change here in Alberta, the time windows in which the skydivers could safely jump were few and far between. We pulled through regardless of these adversities.”

When the day of the jump came, the skydivers jumped five times — as many jumps as the weather would permit. “Theoretically, one jump is enough,” Casiano explained. “But as engineers, we always want to have more data to work with.”

2011 wingsuit jump setup. (Image: NovAtel)

2011 wingsuit jump setup. (Image: NovAtel)

2019 wingsuit jump setup. (Image: NovAtel)

Wingsuit Jumps Compared: Because of the PwrPak7D-E2’s small size yet strong processing power, Levson required fewer devices in 2019 than in 2011, when he was equipped with two receivers, two antennas, a laptop and a battery. The amount of positioning data also increased. (Image: NovAtel)

High-Flying Maneuvers

The skydivers executed four maneuvers during their jumps.

<strong>DART:</strong> This simple jump established a baseline for more complex maneuvers to follow. (Photo: NovAtel)

DART: This simple jump established a baseline for more complex maneuvers to follow. (Photo: NovAtel)

Dart. The skydivers first performed a straight jump, which the team called the Dart. The data from this jump provided a baseline for analyzing the positioning and attitude data.

“This was more important for the attitude analysis, as we have never collected inertial data in a skydiving jump before,” Fulton said.

S-Turn: One of three completed maneuvers. (Image: NovAtel)

S-Turn: One of three completed maneuvers. (Image: NovAtel)

S-Turn. Next came the S-Turn. In this maneuver, Levson weaved from side-to-side to test how the equipment handles agile movements.

For the S-Turn, the engineers anticipated seeing the biggest changes in roll. “We were pleasantly surprised to see that the S-Turn is detectable in the azimuth data as well, indicating high correlation between roll and azimuth in a skydiver’s movements,” Fulton said.

The maneuver revealed that when Levson rolls, his body is using less surface area for wind resistance. As a result, he was falling to the ground faster, which then meant the dataset is shorter.

“This became another challenge during data processing, as the free-fall portion of the datasets were now becoming less than 3 minutes in duration,” Fulton said.
Data from the S-Turn also revealed the effect of crosswinds, which is detectable in the data.

Reverse Immelmann: How the intricate maneuver works. (Image: NovAtel)

Reverse Immelmann: How the intricate maneuver works. (Image: NovAtel)

Reverse Immelmann. The third maneuver was the Reverse Immelmann. Levson flipped onto his back, began a downward turn until perpendicular to the ground, then leveled off, traveling in the opposite direction from where he began.

This complicated exercise provided data for all aspects of an attitude solution — roll, pitch and azimuth. By comparing the expected and real data, the team found several places where the maneuver wasn’t performed perfectly.

“There are many challenges once in the air that would have caused Levson to deviate from the trends in the data that we expected,” explained Fulton. “This is where we realized that our solution was working much more to evaluate the skydiver, rather than using the wingsuit to evaluate our product.”

Casiano agreed. “As a whole, the PwrPak7D-E2 was telling a story about Andrew’s flight,” he said.

The team also wanted to have the skydivers try a Cobra — a maneuver from aerobatics where an airplane momentarily lifts it nose and stalls — but time constraints prohibited it.
“If we had gotten this [a Cobra] recorded, it would have been detectable in the pitch and horizontal velocity data,” Fulton said. “Who knows what other findings we would have come across in this data!”

Measurement matrix: The asterisks (*) denote data values that can only be measured with an IMU. (Chart: NovAtel)

Measurement matrix: The asterisks (*) denote data values that can only be measured with an IMU. (Chart: NovAtel)

Applications

All these tests, of course, are designed to apply to real-world applications where the PwrPak7D-E2’s capabilities are used in dynamic environments.

For instance, an unmanned aerial vehicle (UAV) needs a feedback mechanism that tells the user whether it is moving or hovering. “In the wingsuit project, we proved that crosswind can be detected,” said Casiano. “This is an important finding for UAV applications, since a feedback loop from the PwrPak7 and the SPAN system can help rectify movement from external forces with counter propulsion to stay still. The PwrPak7D-E2 enclosures allow a data rate of up to 200 Hz, meaning you can capture motion with more detail.”

The PwrPak7D-E2 also works well for any black-box application where users want to record with the push of a button.


Inside the PwrPak7D-E2

Photo: NovAtel

Photo: NovAtel

The PwrPak7D-E2 is an all-in-one product. Its components are designed to work together seamlessly to provide positioning data, housed in NovAtel’s OEM7 firmware.

  • GNSS receiver card used to capture positioning data
  • Dual-antenna capability to provide accurate heading
  • Epson IMU to record attitude and motion
  • On-board logging to eliminate the need for constant monitoring on a PC

Post-Processing

Preparation enabled the team to process the data on site. The on-board logging feature on the PwrPak7D-E2 eliminated the need for constant monitoring during data collection. The unit is pre-configured so that at the time of the jump, Levson only needed to push a button for the unit to start collecting data.

Once the pair of skydivers landed, the ground team offloaded the data for processing, similar to using a memory stick, and moved it to a laptop computer.

“We pulled raw measurement data from the receiver and processed those measurements into position and attitude information,” Fulton said.

It took about 30 minutes to determine whether the dataset was viable. Later processing back in the office generated the charts such as those below.

<strong>Expectation </strong> For both the S-Turn and Reverse Immelman maneuvers, a simulated plot was generated at the office to better understand the inertial data produced from the actual wingsuit jumps. (Chart: NovAtel)

Expectation: For both the S-Turn and Reverse Immelman maneuvers, a simulated plot was generated at the office to better understand the inertial data produced from the actual wingsuit jumps. (Chart: NovAtel)

<strong>Reality:</strong> This chart shows the actual data. (Chart: NovAatel)

Reality: This chart shows the actual data. (Chart: NovAatel)

Dynamic Environments

Photo: NovAtel

Photo: NovAtel

The PwrPak7 series can be used in many environments in the automotive, agriculture, marine, defense and UAV fields.

“We are constantly trying to find ways to apply this product to other applications and industries,” Fulton said. “With more testing, we keep finding that the PwrPak7 can be used to solve more challenges.

“We want to push the boundaries of our products. True innovation comes from challenging yourself and hovering outside your comfort zone,” Fulton said. “For this project, we are more than satisfied with the results we found. In order to further challenge ourselves and this product, we look forward to applying the PwrPak7 in more scenarios.”

“The PwrPak7 is a robust unit that sets us up for more exploration,” Casiano said. “We are always looking for more challenges to put this unit through to see how the PwrPak7 can further help solve our customer’s problems.

But will there be more skydiving for NovAtel in Levson’s future?

“We could always revisit the skydiving project in another nine years,” Casiano said. “But who knows how the technology will evolve by then?”


Post flight: Blair Egan (right) and Andrew Levson back on Earth. (Photo: NovAtel)

Post flight: Blair Egan (right) and Andrew Levson back on Earth. (Photo: NovAtel)

What it feels like to take the plunge

For those of us who have never jumped out of a plane, engineer and skydiver Andrew Levson provides insight.

“It’s not as scary as people think. Because the plane is moving fast, it’s mostly just windy and loud. You don’t get that roller coaster type feeling; in fact you don’t feel like you are even falling — freefall feels more like floating than falling. You definitely wouldn’t know you are flying at speeds over 100 mph.

“When you are climbing out of a plane, there is nothing else on your mind aside from the jump you are about to do. It is pure freedom, and there is often no stress, just a sense of peace and an intense focus on your plan for the jump. Once you get out of the aircraft, you get to fly your body in the way that you want to — most people only know of the position of falling with your body arched and belly toward the ground, but there are many different ways you can orient your body. Some of the lesser known ways to fly your body include your arms and legs spread out while flying a wingsuit (with your belly or back toward the earth) or flying with your head pointing straight at the ground.

“When you skydive, you get to explore the sky with your friends, which is an amazing and unique experience. During a skydive, it is common to experience an ultra-focus during the jump — time slows down a bit and you can see and feel things that are seemingly beyond your typical capability.Many people are amazed at how much skydivers are able to do in the short period of time that a single skydive lasts — about a minute for regular skydives and about two or three minutes when flying a wingsuit.”

About the Author: Tracy Cozzens

Senior Editor Tracy Cozzens joined GPS World magazine in 2006. She also is editor of GPS World’s newsletters and the sister website Geospatial Solutions. She has worked in government, for non-profits, and in corporate communications, editing a variety of publications for audiences ranging from federal government contractors to teachers.