Orion spacecraft's avionics designed for reliability in deep space

8Unlike those used in manned spaceflight platforms of the past, the avionics and other electronics used in NASA's Orion spacecraft are driven mostly by software and commercial processor technology ruggedized and radiation-hardened to endure extreme radiation and temperature fluctuations. Orion's updated avionics also can handle the severe acoustic and vibration environments associated with launch, orbit, a fiery reentry, and a saltwater landing.

The spacecraft, named after the constellation it resembles, is ushering in a new era of human space exploration. aims to use Orion to transport humans to an asteroid by 2025, and then to Mars in the 2030s.

While Orion is similar in shape and size to its Apollo-era predecessors, its design, technology, and capability are light-years beyond those early platforms as well as the space shuttle fleet that Orion will be replacing.

During development and production, NASA's Orion avionics team faced numerous design challenges, including building a spacecraft that would be exposed to an extremely inhospitable deep-space environment, as it would be flying through the Van Allen belts (see sidebar). These efforts culminated late last year, when Orion's avionics capabilities were demonstrated during its first flight test (EFT-1), where the spacecraft was equipped with 1,200 for a two-orbit flight that lasted 4.5 hours.

Sidebar 1
(Click graphic to zoom by 2.2x)

Orion's avionics system consists of six main subsystems: Command and Data Handling (C&DH); Guidance, Navigation and Control (GN&C); Communications and Tracking (C&T); Displays and Controls (D&C); Instrumentation; and Power.

"Redundant components are provided for the majority of subsystems," says Paul Anderson, Orion avionics director for Lockheed Martin. "The overall hardware and software integration of Orion's avionics and network system represented a significant technical challenge." The success of the "avionics, software system, and subsequent EFT-1 flight can be attributed to the tremendous amount of integration and testing that was performed in multiple development labs," Anderson says. "This team included numerous NASA and industry partners located across the U.S." (See Figure 1.)

Figure 1: Spacesuited engineers demonstrate how four crew members would be arranged for launch inside the Orion spacecraft, using a mockup of the vehicle at Johnson Space Center. (Photo credit: NASA/Robert Markowitz.)
(Click graphic to zoom by 1.9x)

Computers "fly" the spacecraft

For EFT-1, Orion relied on two flight computers. For its next flight, which is slated for 21 days, the vehicle will have four flight computers aboard – all built by Honeywell engineers and powered by the IBM PowerPC 750 FX processor chip, which was introduced more than a decade ago, in 2002.

Interestingly, "all four flight computers ‘think' they're flying the vehicle," says Matthew Lemke, Orion's manager of avionics, power, and wiring for NASA; Lemke is essentially responsible for almost everything that generates or uses electrons on the spacecraft. "It's not like the shuttle, which used four computers that all voted on the output and had to be in agreement to fly the vehicle. In this case, each computer thinks it's in charge of the vehicle, independently, sending commands. Our power and data units (PDUs) have a priority table that says ‘If I get a command from Computer 1, I'm going to use it. If for some reason it's not there, I'll use Computer 2.' This is much different than programs that rely on a voting architecture."

"The flight computers are a derivative of the computer Honeywell built for the Boeing 787," Lemke says. "We took their commercial processors and ruggedized them for the environment of space, but the basic architecture is the same."

This computer is an example of commercial off-the-shelf (COTS) use in Orion, Anderson points out. This approach "is based on a commercial aircraft computer architecture that incorporates a 750FX microprocessor," Anderson adds. COTS technologies are playing "a selective but critical role throughout Orion," he notes.

The reason for this approach: COTS parts can provide affordable design alternatives through shorter development schedules at a much lower cost than fully qualified components. "Wherever possible, we try to leverage what's available commercially or from defense to help reduce our costs," Lemke says.

COTS or not, before any components can be used on Orion, they must first undergo an extensive program that includes "radiation screening, vibration and shock testing, thermal vacuum testing, and electromagnetic compatibility," he notes.

Although individual COTS processors can be susceptible to radiation upsets, Orion's overall architecture relies on self-checking pairs of processors to minimize system effects of radiation upsets. "Orion also uses COTS hardware in several non-mission-critical areas within development of flight instrumentation and video subsystems," Anderson says.

While the use of commercial components helped reduce costs, Orion achieved huge savings in mass and volume by combining typically separated functions within a single component – the PDU. "This saved overall spacecraft mass and volume significantly by performing command and data handling, power switching, pyrotechnic, network switching, instrumentation collection, and environmental control functions within integrated units," Anderson says.

Software drives Orion avionics

How important is software to Orion? "It's everything," Lemke says. "Think back to Apollo: every system from propulsion to life support to displays to power – each was a self-contained system, run with something very simple to control."

Now, on Orion, NASA engineers use flight computers and "boxes called PDUs, and everything on the vehicle connects in via the PDUs to the flight computers," Lemke elaborates. "Very little isn't under computer control. We have few switches – in the neighborhood of 40 on the vehicle total – compared with an airline or even the shuttle cockpit with all the switches."

Orion uses a Federal Aviation Administration (FAA)-certified operating system for and high criticality applications. This is to ensure that "one part of the software can't affect another part of the software. The whole goal is reliability," Lemke says.

New tech for Orion's data network

For its data network, NASA experts are leveraging 1 GB/s time-triggered Gigabit Ethernet (TTGbE) network, which was developed by Austrian company TTTech, who worked with Honeywell to build the system commercially. Lockheed Martin and NASA then joined in to make it work for Orion.

"The Orion Data Network (ODN) is the first space-based implementation of TTGbE technology and provides a deterministic, synchronous, and congestion-free Ethernet network communication," Anderson says.

It's essentially standard Ethernet, but with protocols layered on top to guarantee timing and delivery. The system is designed "to run all of our data over three separate links so that if any of the network switches or cables or anything else experiences a problem…as long as one of those three get across, the data is all there and it will work fine," Lemke points out. That's a level of redundancy you don't normally find in networks.

TTGbE also enables hosting asynchronous Ethernet traffic without interference to critical deterministic traffic. "The deterministic nature of Orion's ODN guarantees timing of messages traveling through the network, allowing us to integrate a complex suite of redundant main flight computers, backup flight computers, guidance and navigation sensors, radios for off-board communication, crew display and command interfaces, and primary and developmental instrumentation systems," Anderson notes.

"In space, we want to know if the computer issues a command to a thruster and that it really happens right then and has the highest priority," Lemke adds. "So in TTGbE, we guarantee commands get across and data goes exactly when it should. Then, with functions like video or audio, it fills in the free time on the data network. It's ‘best-effort traffic,' which is guaranteed delivery." The TTGbE technology has now become an industry standard – SAE AS6802 – and "hopefully others will begin using it as well," he says.

Avionics survive splashdown

Orion's flight computers, software, and data network all did their job during the launch and flight in orbit, but NASA also wanted to make sure the electronics survived the splashdown. Upon returning from space, Orion makes saltwater landings with the help of specially designed parachutes to slow its descent, so the avionics design team had yet another extreme environment to factor in. To handle it, "all of our connectors and harnesses or enclosures are designed to be able to handle immersion in saltwater," Lemke notes.

During Orion's EFT-1, its highly complex avionics system "performed near flawlessly," says Anderson. "Very few anomalies were experienced across the entire mission. The stability of Orion's data network was demonstrated by ‘zero' bit errors during flight – despite operating and flying through extremely challenging radiation and thermal environments," Anderson adds. (See Figure 2 on following page.)

Figure 2: Recovery-team members in rigid-hulled inflatable boats approach NASA's Orion spacecraft following its splashdown in the Pacific Ocean. A combined team from NASA, the Navy, and Orion prime contractor Lockheed Martin retrieved it for return to shore. (Photo courtesy of U.S. Navy.)
(Click graphic to zoom by 1.9x)

Orion's imaging system also provided spectacular, high-resolution video of key mission events – including capturing the service module fairing jettison, launch abort system separation, forward bay cover release, and drogue and main parachute deployments for the water landing.

"The video showed incredible live views of Earth from 3,600 miles while Orion traveled through the Van Allen radiation belt," Anderson says. "This unique video imagery also captured the multicolor plasma trails that occurred during the 20,000 mph reentry."

Orion's next flight will involve a journey beyond the orbit of the moon, with the goal of getting NASA within one step of being able to fly humans to an asteroid or Mars.

"Our EFT-1 test flight was a big flight to prove out our heat shield and parachutes and that all of it would work during the mission – our highest risk areas," Lemke says. "This next flight is to test out all of the other systems on the vehicle to ensure it's one step away from putting an astronaut on it." Now NASA will start putting on all of the systems that may not have been on the first vehicle.

"In terms of avionics, we're so excited that NASA had a successful first launch and we're now able to move toward getting people into space," says Lemke. "We're shooting for our next flight to be at the end of 2018, using the Space Launch System being built at the Marshall Space Flight Center."