New technique fusing GPS and camera data could help satellites navigate autonomously above low Earth orbit

April 4, 2021
New technology could let spacecraft dodge satellites in orbit. (Unsplash/NASA)

New technology could let spacecraft dodge satellites in orbit. (Unsplash/NASA)

Researchers are seeking a patent for a technique that would estimate the relative locations of satellites beyond low Earth orbit with unprecedented accuracy, which could allow spacecraft to navigate autonomously at high altitudes.

The new method combines signals from GPS satellites with observations from cameras or other sensors to determine where one spacecraft is in relation to another, which could aid in satellite maintenance and allow for missions involving multiple spacecraft. The patent application for the technique was published by the U.S. Patent and Trademark Office on Feb. 18. 

"It's a strategy that is meant to use something which already exists," said Vincenzo Capuano, a visiting researcher at the California Institute of Technology and co-inventor of the process. "We are able to basically get higher accuracy than we would using just GPS or using just vision."

On Earth, GPS navigation is relatively straightforward. A device, such as a smartphone, receives a radio signal broadcast by a satellite belonging to the U.S.' Global Positioning System, or another kind of Global Navigation Satellite System, or GNSS. Because the smartphone "knows" how fast the signal can travel and the satellite's precise location, it can calculate how far away the satellite is. Knowing the distance between itself and at least four satellites allows the smartphone to determine its own location. 

GPS satellites orbit around 20,000 kilometers above the ground, within a zone known as medium Earth orbit. 

"They are distributed in such a way as to create a constellation which enables full coverage of Earth's surface," Capuano said. 

These satellites were designed for navigation by land, sea and air, so their antennas are pointed at the planet's surface. In the last decade, though, interest has grown in using GNSS satellites to help other spacecraft navigate, Capuano says. This has already been successfully accomplished for spacecraft in low Earth orbit, which extends to altitudes of around 1,000 kilometers.  

At higher orbits, where weather and communications satellites are often found, the signals become weaker and difficult to process. 

"If you want to use a GPS signal in an orbit which is above the orbit of the GPS satellite, then it starts to become very challenging," Capuano said. 

At these altitudes, spacecraft currently use cameras and lidar, or Light Detection and Ranging, to determine their relative positions. However, this method typically is only accurate to within a few centimeters. 

"The problem that I'm trying to address is not just spacecraft location but [also] autonomous spacecraft navigation," Capuano said. "We want to make the spacecraft completely independent ... so it is able to determine its location without any information coming from the ground [control center]."  

It can take $400 million to set up and launch a satellite into geostationary orbit, an altitude of around 36,000 kilometers, Capuano says. Autonomous spacecraft could extend the useful life of an aging satellite by providing inspections, repairs and refueling. 

"You have one spacecraft approaching another spacecraft and they are both orbiting very fast and you cannot wait for commands from [the] ground," he said. "So the spacecraft ... has to be completely autonomous; it has to be able to modify its trajectory and to determine its position with respect to the other spacecraft." 

Additionally, this would open up possibilities for missions executed by multiple spacecraft working together in tight formations. 

To enable spacecraft to determine their relative positions, Capuano and his colleagues developed an approach that integrates raw data from several different sensors. Two spacecraft might measure the distances between themselves and a number of GNSS satellites below them, compare these measurements and use the difference to provide an initial rough estimate of how far apart they are from each other. This estimate would then be supplemented by another from the cameras on each spacecraft.

"It's basically fusing this GPS observation together with images acquired onboard," Capuano said. He and his colleagues calculate that their algorithms can estimate the relative positions of satellites in a geostationary orbit to an accuracy of about 2 millimeters. However, Capuano notes, the technique may become less accurate at higher orbits. 

The method could also be adapted for other equipment besides cameras, such as radios or lidar sensors, he added. He estimates that it may be ready for practical use in about three to five years. 

As an initial step to validate the strategy, Capuano and his colleagues will test its performance under terrestrial conditions that approximate the challenging environment above the constellation of GPS satellites, including by collecting signals in the presence of tall buildings. 

The NASA Jet Propulsion Laboratory provided funding for the research and the government has "certain rights in the invention," the patent application states.

The application for the patent, "Systems and Methods for Robust and Accurate Relative Navigation," was filed Aug. 17, 2020, to the U.S. Patent and Trademark Office. It was published Feb. 18, 2021 with the application number US16/995652. The earliest priority date was Aug. 16, 2019. The inventors of the pending patent are Vincenzo Capuano and Alexei Harvard, California Institute of Technology; and Soon-Jo Chung, California Institute of Technology and NASA Jet Propulsion Laboratory. The assignee is California Institute of Technology.

Parola Analytics provided technical research for this story. 

We use cookies to improve your experience on our site and to show you relevant advertising.