simulating
REALITY MSC Software Magazine
recognized the importance of simulation from the very beginning and devoted an average of three people working with multibody dynamics over the life of the project. The rover, the most critical part of the simulation, was modeled to a high level of fidelity including many flexible elements some of which incorporated nonlinear stiffness and damping. The descent stage model is much simpler, consisting entirely of rigid bodies. In the beginning of the project, separate models were used for the rover separation, mobility deployment, and touchdown phases. During the later stages of the project, all of the models were emerged into one. The combined model runs between 17 to 93 minutes on a four-CPU Hewlett Packard Unix workstation.
Determining Loads for Structural Design Optimizing the design of every component in the payload was critical because of the need to ensure their ability to withstand loads and successfully perform their mission while at the same time minimizing their weight because of the high cost of transporting incremental weight to Mars. Adams was used to predict the loads on components and subassemblies and these loads in turn were used as input to structural analysis that optimized the design to provide the strength to withstand mission loads while minimizing size and weight. The philosophy of the modeling was not to try and predict every event to 100% accuracy but rather to determine the bounding limit design loads that could be expected on every component. Prior to the Adams simulation, JPL engineers believed that the projected 1 meter per second maximum speed during touchdown would
not induce particularly high loads on the rover. However, simulation showed the loads were much higher than expected. The original expectations also were that the rover would be in a quiescent state when it touched down on the Martian surface but the Adams simulation showed that the rover was actually rotating and swinging as it landed. As a result, the rover structure was stiffened to mitigate these issues. Later studies provided the additional surprise that the deployment of the rover’s wheels and struts as originally planned generated even greater loads on some of the rover components than the touchdown. Simulation showed that the end of wheel deployment generated hammer-like blows on the rover suspension and frame.
Volume II | Supplemental Issue - Summer 2012
solver. This made it possible to validate the system performance and tune the controller parameters with a detailed mechanical model. JPL engineers validated and updated the Adams model by correlating the simulation results to the test data.
Only One Chance to Get it Right Adams Simulations Help Curiosity Rover Make Perfect Touchdown on Mars
The engineers at JPL were not able to test most of the critical mission events on Earth so they had to rely upon simulation to design most of the critical hardware and control sequences for this mission. The accuracy and thoroughness of the simulation results helped make it possible to successfully and precisely place the rover onto the Red Planet. u
ONLY ONE CHANCE TO GET IT RIGHT
Image Credit: NASA/JPL-Caltech
JPL engineers addressed this problem by changing the timing of the wheels and struts deployment. Changing the timing of the wheels and struts deployment also reduced the swing rate and swing angle before touchdown.
SAVE THE DATE
Checking Other Aspects of the Landing Sequence
2013 USER CONFERENCES
DATES & LOCATIONS Americas
Tuesday / Wednesday - May 7/8
EMEA
The most critical aspect of the separation between the rover and descent stage is the need to avoid contact between the flight hardware. The Adams simulation was used to check the clearance and ensure that there was no possibility of contact. In the final design there is very small clearance but no contact issues.
The events will take place in May and June. Location details coming soon.
Sweden - Monday / Tuesday - May 13/14 Germany - Tuesday / Wednesday - May 14/15 UK - Wednesday / Thursday - May 15/16 France - Thursday / Friday - May 16/17 Russia - Tuesday / Wednesday - May 21/22 Italy - Wednesday / Thursday - May 22/23 Turkey - Monday - May 27th
*Dates and locations are subject to change.
Japan
MSC Software will celebrate its 50th anniversary in February, 2013. Please join us in celebrating at the 2013 User Conferences.
Thursday - May 30th
The flight control software was written in C++. The people who wrote the controller software required a detailed mechanical model to accurately predict the system performance. The engineers overcame this issue by compiling the controller with the Adams
APAC
For more information, visit: www.mscsoftware.com/50years
Volume II | Supplemental Issue - Summer 2012
FEATURE STORY
| 2
Korea - Monday / Tuesday - June 4/5 China - Wednesday / Thursday - June 6/7 India - Thursday / Friday - September 5/6
ADAMS SIMULATIONS HELP CURIOSITY ROVER MAKE PERFECT TOUCHDOWN ON MARS
Based on an interview with Dr. Chia-Yen Peng, Principal & Lead Engineer for the Loads and Dynamic Simulation team at JPL.
M
ars Science Laboratory is a robotic space probe mission that successfully landed Curiosity, a Mars rover, in Gale Crater on Mars on August 5th, 2012. The sky crane landing sequence required the rover to transform from its stowed flight configuration to landing configuration while being lowered to Mars wheels down from the descent stage. The final entry, descent, and sky crane landing phase was called “seven minutes of terror” by NASA engineers because of its complexity and the fact that human intervention from earth was impossible. NASA Jet Propulsion Laboratory (JPL) engineers simulated this final sky crane landing sequence using MSC Software’s Adams multibody dynamics software. The simulation identified problems with the initial concept design and guided engineers as they resolved these issues and made the design more robust. The simulation was also used to validate the landing sequence and determine loads on subassemblies and components. The controls software code that guides the mission through the sky crane landing sequence was integrated into the Adams environment to validate and tune its performance. The accuracy of these simulations was proven by the success of the mission.
1 | MSC Software
Complex Landing Sequence The Curiosity rover is about twice as long and five times as heavy as previous Mars Spirit and Opportunity rovers. Landing such a large payload on Mars is challenging because the atmosphere on Mars is too thin for parachutes and aerobraking to be effective yet still thick enough to create the potential for stability and impingement problems when decelerating with rockets. Also at 900 kg the Curiosity rover is too heavy to use airbags to cushion the shock of landing. One additional challenge is that the 14 minute trip required for a radio signal to travel from Mars to earth means that the vehicle must act autonomously without interactive control from Earth for the entire landing sequence. NASA engineers addressed these challenges by developing a unique entry-descent-landing system. The aeroshell containing the heat shield, rover, descent stage, backshell and parachute, separated from the cruise stage ten minutes before entering the Martian atmosphere and fired thrusters to orient the heat shield facing Mars. The aeroshell slowed to about 578 meters per second at about 10 kilometers from Mars at which point a supersonic parachute deployed.
At about 1.8 kilometers altitude with the aeroshell traveling at 100 meters per second, the rover and descent stage dropped out. The descent stage fired its rocket thruster to slow to less than 1 meter per second and lowered the rover with a 7.6 meter tether consisting of three bridles and an umbilical cable carrying electrical signals to the Martian surface. As the bridles unreeled, the rover’s six motorized wheels snapped into position to prepare for landing. After landing, the rover fired explosive devices activating cable cutters to free itself from the descent stage and the descent stage crash landed hundreds of meters from the rover. At 6:33 a.m., the rover confirmed a successful landing.
Only One Chance to Get it Right There was only one chance to get it right. The complex sequence either works perfectly or the entire mission goes up in a cloud of Martian dust as the rover is destroyed. Engineers cannot test the landing sequence on Earth because they can’t duplicate Martian gravity, atmosphere, and the landing conditions on Earth. They can test individual components but the only way to test the complete sequence and determine the loads on the individual components is with simulation. The program engineers
2
START OF LANDING SEQUENCE 1
Curiosity and its landing spacecraft, traveling at 13,200 miles per hour, had just seven minutes to slow down to two miles an hour and land safely on the Martian surface, a remarkable engineering achievement considering that the complex maneuvers and sequences all had to work to perfection, the first time tried!
ENTRY. DESCENT. LANDING.
turn to entry
1
Curiosity and its landing spacecraft were modeled extensively with MSC’s Adams software, to a high level of fidelity, including many flexible elements some of which incorporated nonlinear stiffness and damping.
start of landing sequence
CRUISE TURN TO ENTRY 2
Nine minutes before entry, the back shell thrusters oriented the spacecraft so the heat shield faced forward, a maneuver called “turn to entry.” After the turn to entry, the back shell jettisoned two solid-tungsten weights, called the “cruise balance mass devices.” Ejecting these devices, which weigh about 165 pounds (75 kilograms) each, shifted the center of mass of the spacecraft. This allowed the spacecraft to fly through the atmosphere at an angle, generating lift. The lift was used for “guided entry” to steer out of unpredictable variations in the atmosphere and improve landing precision.
GUIDED ENTRY 3
As the spacecraft initially interacted with the upper atmosphere, thrusters on the back shell adjusted the angle and direction of tilt, letting the spacecraft fly a series of “S” curves. These curves reduced the horizontal distances the spacecraft covered as it descended. The guided entry maneuvers also corrected for drift to the left or right due to winds.
PEAK HEATING 4
Eighty seconds after entering the atmosphere, the heat shield reached the hottest temperature ever for any Mars entry vehicle, about 3,800 degrees Fahrenheit (2,100 degrees Celsius). The innovative Phenolic Impregnated Carbon Ablator (PICA) thermal protection system, allowed Curiosity to remain at a cozy temperature of 32 to 77 degrees Fahrenheit (0-25 degrees Celsius) inside its aeroshell. The PICA thermal protection materials were subjected to severe thermal and mechanical loading environments during entry into martian atmosphere. MSC Software’s Marc was used for the Thermal-Structural Analysis of the PICA Tiles.
HYPERSONIC AERO-MANEUVERING
5
The 7,000 pound spacecraft, traveling at Mach 5, caused the molecules in the atmosphere to break apart, creating a layer of hot plasma. Shockwaves caused air density and pressure to waver dramatically. Onboard guidance algorithms controlled the direction of lift vector, via bank angle modulation, to keep the vehicle on the desired trajectory. Specifically, the entry flight path angle had to be maintained within a strict margin of only 0.27 deg as the spacecraft slowed to under Mach 3. This extreme level of flight accuracy was accomplished with the help of Adams. The flight control software, written in C++, required a detailed mechanical model to accurately predict the system performance. The engineers overcame this issue by compiling the controller with the Adams solver. This made it possible to validate the system performance and tune the controller parameters with a detailed mechanical model. JPL engineers validated and updated the Adams model by correlating the simulation results to the test data.
PARACHUTE DEPLOY 6
7 miles (11 kilometers) above the Martian surface and traveling at Mach 2, the parachute was deployed. This was the largest parachute ever built for a planetary mission to land on the Red Planet. The 64.7 foot (19.7 meter) disk-gap-band style chute was deployed by a mortar. The main disk was a dome-shaped canopy with a hole in the top to relieve the air pressure. A gap below the main canopy allowed air to vent out to prevent the canopy from rupturing. During its use, it generated up to 65,000 pounds (almost 29,500 kilograms) of drag force and slowed the spacecraft down to 250mph.
HEATSHIELD SEPARATION 7
Adams Simulations Help Curiosity Rover Make Perfect Touchdown on Mars 3
4 guided entry
RADAR DATA COLLECTION peak heating
8
CURIOSITY Curiosity was modeled in Adams to a high level of fidelity including many flexible elements some of which incorporated nonlinear stiffness and damping. In the beginning of the project, separate models were used for the rover separation, mobility deployment, and touchdown phases but during the later stages of the project all of the models were emerged into one. The combined model runs in between 17 to 93 minutes on a Hewlett Packard Unix workstation with four central processor units (CPUs).
At 5 miles (8 kilometers) above the surface of Mars, the heat shield separated from the spaceraft and plummeted to the ground. With the heat shield off, the Mars Descent Imager started recording the first ever video of a Mars landing. This movie gave us a “rover’s eye” view of the rapidly approaching Martian surface as if we were landing with it. To see this video, go to: www.mscsoftware.com/roverview.
As Curiosity descended, six independent pulsed-Doppler radar antennas where used to scan the surface. Distance readings from this radar system were used to signal when the parachute should be detached, when retro-rockets should fire, and when the rover should be lowered down from the descent stage. Engineers were able to successfully “lock-on-target” by the time the rover was about 1.2 miles (3 kilometers) above the surface. While the radar only functioned for two minutes, it was so important that engineers tested it over 100 hours, dangling it from a helicopter over Mojave and Death Valley.
6 BACKSHELL SEPARATION AND POWERED DESCENT
parachute deploy
5
9
8
hypersonic aero-maneuvering
radar data collection
At this stage, the spacecraft had to perform some of the most technically challenging maneuvers ever accomplished on a space mission. Luckily it was not a total “Leap of Faith” as the sequences were first analyzed with MSC’s Adams multibody dynamics simulation software. For instance, one of the most critical aspects of the separation between the rover and descent stage is the need to avoid contact between the flight hardware components. Adams simulation was used to check the clearance and ensure that there was no possibility of contact during the sometimes violent motions. In the final design there was very small clearance but no contact issues.
SKY CRANE MANEUVER 7 heatshield separation
10
10 sky crane maneuver
ADAMS
DESIGN INSIGHT
Adams was used to predict the loads on components and sub-assemblies and these loads in turn were used as input to structural analysis that optimized the design to provide the strength to withstand mission loads while minimizing size and weight. The philosophy of the modeling was not to try and predict every event to 100% accuracy but rather to determine the bounding limit design loads that could be expected on every component.
JPL Engineers dealt with complexities involving Martian gravity, atmosphere, surface slope, and landing velocities that could not be duplicated exactly here on Earth, and relied on the simulations to gain the insight they needed to feel confident in the execution of the mission. The series of Adams simulations took place in parallel with design – but it was insight from the simulations that helped guide the design to maturity, and to prevent any failures resulting from potentially harsh loading conditions during the mission.
9
12 separation & powered descent
TOUCHDOWN
flyaway
SKY CRANE MANEUVER
11 11
One Body Phase
(Vertical Descent)
(DRL /Bridle Deployment)
12 Two Body Phase Two Body Phase (Constant Velocity)
(Touchdown Event)
Fly-Away Phase
As Curiosity was gently lowered, it eventually made contact with the Martian surface. Prior to the Adams simulation, JPL engineers believed that the projected 1 meter per second maximum speed during touchdown would not induce particularly high loads on the rover. However, simulations showed the loads were much higher than expected. The original expectations also were that the rover would be in a quiescent state when it touched down on the Martian surface but the Adams simulation showed that the rover was actually rotating and swinging as it landed. As a result, the rover structure was stiffened to mitigate these issues.
FLYAWAY
touchdown
Two Body Phase
At an altitude of about 66 feet (about 20 meters) above the surface, the spacecraft then performed the now infamous “Sky Crane Maneuver” - requiring the rover to transform from its stowed flight configuration to landing configuration while being lowered to Mars wheels down from the descent stage with suspension cables. JPL Engineers used Adams to predict the dynamics of the separation and lowering sequences. For instance, as the bridles unreeled and the rover’s six motorized wheels snapped into position to prepare for landing, Adams simulations showed this “snapping” effect generated even greater loads on some of the rover components than the touchdown. The simulation results showed that, at the end of wheel deployment, severe hammer-like blows where generated on the rover suspension and frame. JPL engineers addressed this problem by changing the timing of the wheels and struts deployment. Changing the timing of the wheels and struts deployment also reduced the swing rate and swing angle before touchdown.
Once the descent stage detected touchdown, pyrotechnic charges cut the bridle and “umbilical cord.” With this final directive and release, the descent stage no longer had access to the rover’s computer “brains.” Having completed its mission and now “unconscious,” the descent stage flew out of the way, coming to rest on Mars several hundred yards from the Curiosity.