NASA Selects Advanced Robotics Projects for Development
NASA has a long history of developing cutting-edge robotic systems for use in space exploration. NASA also partners with American businesses, universities and other federal agencies to transfer those technologies back into the nation's industrial base, improving manufacturing capabilities and economic competitiveness.
Recently, tremendous advances in robotics technology have enabled a new generation of assistive systems and devices in industries as diverse as manufacturing, logistics, medicine, health care, military, agriculture, and consumer products.
As part of the National Robotics Initiative, NSF, NASA, the National Institutes of Health and the U.S. Department of Agriculture have managed a joint solicitation, seeking to engage our next generation of roboticists for the new global technology economy. All participating federal agencies are working with partners to foster the exchange of ideas and technologies that will directly benefit American today and well into the future.
NASA has selected eight advanced robotics projects that will enable the agency's future missions while supporting the Obama administration's National Robotics Initiative. The projects, ranging from technologies for improving robotic planetary rovers to humanoid robotic systems, will support the development and use of robots for space exploration, as well as by manufacturers and businesses in the United States.
Robots can work beside, or cooperatively, with people to enhance individual human capabilities, performance and safety in space as well as here on Earth. Co-robotics, where robots work cooperatively with people to enhance their individual human capabilities, performance and safety is a valuable tool for maintaining American leadership in aerospace technology and advanced manufacturing.
The proposals NASA has selected for development are:
The National Science Foundation (NSF) managed the solicitation and peer review selection process for these NASA awards. Awards range from $150,000 to $1 million, with a total NASA investment of $2.7 million.
(Link: https://www.nasa.gov/home/hqnews/2012/sep/HQ_12-323_NASA_NRI_Advanced_Robotics.html)
NASA's Office of the Chief Technologist (OCT) has commissioned a series of technology roadmaps that identify both component level needs and larger challenges and missions for the next 20 years. Of the 14 roadmaps, Technology Area 4 is titled "Robotics, Tele-Robotics and Autonomous Systems". Several challenges are identified that are well aligned with the co-robotics theme for developing machines to help humans explore space:
(Link: https://www.nasa.gov/robotics/index.html)
NASA Holds Final Sample Return Robot Competition
After five years of competition by more than 40 different teams from around the globe, NASA’s Sample Return Robot Challenge has reached its final stage. The top seven teams competed for the $1.36 million prize purse on the campus of Worcester Polytechnic Institute (WPI) in Worcester, Massachusetts, September 4-6, 2016.
In this final round of the challenge, teams have up to two hours each to locate as many as 10 unknown samples that vary in size, shape, location and difficulty. The samples are classified as easy, intermediate and hard and are assigned corresponding point values. One team could win the entire prize purse, or multiple teams could share a percentage of the prize. Qualifying teams for the final round were:
Prior to this final round of competition, the teams competed in Level 1, where robots had to return two known sample types but from an unknown location within 30 minutes without human control or the aid of Earth-based technologies, such as GPS or magnetic compassing. Since the challenge began in 2012, only seven teams have advanced to Level 2.
The Sample Return Robot Challenge, part of NASA’s Centennial Challenges Program, aims to encourage innovation in robotics technologies relevant to space exploration and broader applications that benefit life on Earth. This event brings together tech-savvy citizens, entrepreneurs, educators and students to demonstrate robots that can locate and collect geologic samples from a wide and varied landscape without human control and within a specified time.
NASA’s Centennial Challenges program is part of the agency’s Space Technology Mission Directorate (STMD). STMD uses challenges to gather the best and brightest minds in academia, industry and government to drive innovation and enable solutions in important technology focus areas. WPI has hosted the Sample Return Robot Challenge since it began in 2012.
(Link: https://www.nasa.gov/press-release/nasa-holds-final-sample-return-robot-competition)
2015-16 Challenge Frequently Asked Questions (FAQ)
Below are FAQ regarding the Challenge Rules. Team Leaders of registered teams could submit questions about the rules by emailing challenge@wpi.edu. All new questions weree posted and answered in the online FAQ at http://challenge.wpi.edu.
F1. Can you describe in more detail how the prize money could be distributed?
Yes.
For Level 1:
All teams who successfully complete Level 1 will split $50,000, with a maximum of $5,000 per team.
For Level 2:
Below are some specific examples:
Example 1:
The top three teams score 10, 9, and 5 points respectively. Since the first place team scored 10 points, the total available to be distributed is $750,000 (see P8).
Example 2:
The top 2 teams score 4 and 5 points. No other teams score points in Level 2. Since the first place team scored 5 points, $250,000 is available to be distributed.
Example 3:
F2. Will false samples be placed on the field?
No. We will not intentionally place any false samples on the field of play. For example, we would not place an out-of-spec tennis ball on the course of play and we will scan the entire course for debris that could potentially be misinterpreted as a sample prior to the start of the challenge attempts. However, if your robot collects an item that it though was a sample that isn’t, it will count towards your non-sample mass.
F3. Will all the samples be on the course for Level 1?
No. The Level 1 course contains only the PCS and one easy sample.
F4. Will we have access to the samples once we arrive on-site?
Yes and no. Teams will be given some access to samples in the robot pit area with the following restrictions. For the Easy samples, all teams will have access to see, feel, touch, and calibrate to the actual samples we will use. For the Intermediate samples, all teams will have access to view the samples from a distance of no more than 15 meters in a controlled area. No teams will have any access to the hard samples until they identify them on the course.
F5. Can you provide more information of the exact nature of known samples or obstacles on the course?
Yes. While we are not suggesting you should purchase the items from the following retailers or this is the only place from which they can be purchased, below are links to the actual items referenced in the rules:
Orange Warning Fence (Field Boundary) – Please note we are showing this as a sample for anyone who may not know what we are describing. We will be seeking donors of this since there is a lot to purchase, so we may not know the final brand until we secure this. If what we end up getting for use is not available in retail quantities, we will do our best to get a sample to each team: http://www.homedepot.com/buy/building-materials/fencing/tenax-guardian-safety-fence/4-ft-x-100-ft-orange-warning-barrier-20640.html
Paint for easy samples: Rust-oleum Painter’s Touch Ultra Cover (Paint + Primer), Gloss Grape 249113. UPC 020066187675
Sample of HSV color scale: http://i.stack.imgur.com/LC8Oh.png (240-60 means the blue, purple, red, orange side)
F6. Is the robot allowed to climb “immovable obstacles” on the terrain?
Yes. However, be aware of R5 if these behaviors have the potential to severely damage the obstacle.
F7. What kind of surfaces can we expect to encounter on the course?
You can expect to encounter firm ground and a variety of walkable surfaces. This would include turf, pavement, packed dirt, short grass, and possibly traversable rocks (i.e. gravel). You are not expected to move through loose mediums like sand, travel through water, or negotiate tall grass.
F8. Will people be allowed within view of the robot’s sensors?
There will be no spectators inside the boundaries of the course or inside the boundary fencing. The only people allowed on the course will be event officials or individuals approved by the event for specific purposes (e.g. filming). Those allowed on the course will be clearly identified by their badge and clothing.
F9. Will the samples be placed on a table or buried in the ground?
No. The best effort will be made to have all samples placed on the surface of the course. Absolutely none will be buried or in water. In some cases, like with the tennis ball, it may be raised very slightly or contained in order to prevent it from rolling far from its location. An example would be placing a small rubber O-ring under the ball for it to sit on, so it is not sitting directly on the ground but extremely close to the surface.
F10. Can we get an unofficial inspection before our official one?
Yes. Any time before your challenge attempt or your robot is impounded your team may request an unofficial inspection. An inspector will review your robot for compliance and attempt to answer any questions you may have. A scale will also be available during this time. While these inspections are not final, our goal is to help make sure that every team that arrives with a robot is compliant with the rules and that does not stop them from competing.
F11. Can I move my robot from the on-deck area to the starting zone for Level 1 by driving it under its own power?
No.
F12. If my team only has one member at the event, will I be able to get assistance to move my robot?
Yes. You can ask for help from other teams or event officials. However, moving the robot is ultimately your responsibility and any damage that may occur during this process is your responsibility.
F13. Is the Home Beacon platform considered part of the starting platform?
No. This means that no part of your robot can start on or overhanging the Home Beacon platform. Additionally, any home beacon components not completely contained within the home beacon platform for the duration of the run will be considered part of the robot. Additionally, any samples that end up on or overhanging the Home Beacon platform will not count.
F14. Are teams allowed to mark the starting platform? How will we know it is ours?
Each starting platform will be painted a bright color and teams will know which platform they are starting on prior to each challenge attempt. In addition, teams are allowed to mark the platform as long as they do not permanently alter the platform and anything used to mark the platform is included in the robot mass, starts within the marked starting area, and violates no other rules. The Home Beacon starts on a separate platform directly behind the starting platform and is designed to aid competitors in this issue.
F15. How important is the separation of the samples from one another?
Obviously, when collecting samples from an unknown area, sterile handling would be extremely important for their scientific evaluation. For the purposes of this challenge, this is an important area but not a critical one we are looking to investigate. For example, teams may employ simpler methods like separate compartments within the same box or wrapping the samples individually and placing them in a single box. Judges will only be looking to ensure that the surfaces of any samples never come in contact with one another.
F16. How do we know if our samples are “easy and obvious to remove”?
The goal with this rule is to ensure that the judges can easily access the sample to determine if they have come in contact with other samples, to analyze the mass of all components returned, and to evaluate whether the samples are within the vertical projection of the starting platform. If a sample is incredibly difficult to access or cannot be accessed without moving the robot, the judges may deem those samples inaccessible and not count them.
Teams will be asked to provide documentation to the inspector that clearly describes how to access where any items are stored within the robot. Accessing these items may require tools, and these must be provided by the team to the inspector.
F17. What will be interpreted as “damage” to a sample?
A sample will be considered damaged if it has a permanent deformation or change in dimension.
F18. When my robot is paused, what exactly needs to stop?
When your robot is paused, it will most likely be done so for the safety of an event official on the course, or to allow another robot to pass in the case of multiple robots. For the safety aspect, it is critical that driving cease as well any outboard motion. It is not expected that your computing or sensing systems shut down, as it would likely be a tremendous time penalty for them to restart. However, any teams that wish to have items that continue moving during a pause must request and be approved for a pause exception. All decisions on what is or is not allowed are at the discretion of the judges.
F19. Can my robot send information to me or a computer outside the course, as long as I am not transmitting any information back?
Absolutely not. There is no communication in any direction allowed with the robot from anything not contained within the course, inspected before the run, and included in the starting size and mass of the robot during Level 1. While we understand this would be only to help you learn the robot’s processes better, everyone has to understand it’s a slippery slope. You are welcome to record data on-board. During Level 2, communication is limited to the Communication Update Periods (section 2.7.2.1)
F20. What does the 80kg mass of the robot apply to?
The 80kg mass applies to everything you as a competitor bring to the event and put on the starting platform to compete as part of your robot. This means it includes batteries, computers, e-stops, safety lights, and anything you might leave behind on the platform or on the course but needs to start with the robot. It does not, however, include the pause switch(es), the required payload in section 1.2, the home beacon, or any samples or materials collected during the run.
F21. Can we use a device onboard that has a GPS, accelerometer, compass, etc, as long as we don’t use those features in our code or our challenge attempt at all?
Yes. We understand that it is tough today to purchase technology that doesn’t include some of these components, even if they will not be used, and therefore we don’t want to make the challenge even more difficult for anyone. If teams utilize devices with any of these disallowed technologies, the onus will be on the team to prove beyond a reasonable doubt that they are not using them during the competition. Teams should be aware that it will be the determination of judges and inspectors as to whether a team has proven compliance with these rules, and teams may be asked to modify or remove certain components to make their robot legal to compete.
F22. Are accelerometers allowed?
Yes, provided they comply with Section 1.3. Be aware that any sensors that utilize magnetic compensation will be disallowed.
F23. Are flying robots allowed?
Provided they comply with Section 1.3 and 1.4 of the rules.
F24. Can we leave objects/beacons/robots on the field at the end of our competition run?
Yes. All items will be removed by event officials at the end of each competition run.
F25. Can we have multiple robots on the field as long as they all start within the specified dimensions?
Yes, see also Section 1.4 of the rules.
F26. Can we use spring-damper systems for shock absorption and suspension?
Yes, provided it is a sealed system and could theoretically work in a vacuum, and complies with Section 1.3 of the rules.
F27. Are spawn allowed to communicate with each other and with the home beacon?
Yes, provided the communication meets all rules on allowed communication, disallowed technology rules, and FCC regulations.
F28. “The required payload may contain a strong magnetic source and frequency jammer to….” Couldn’t this magnetic source directly interfere with R6?
The required payload is designed to aid judges and inspectors in enforcing the rules on allowed and disallowed technologies. Teams are required to submit documentation about their robots, beacons, and communication protocol approximately 6 months prior to the event with additional information on-site. Provided teams submit accurate and reasonable information about their plan and update any changes in a timely fashion, the required payload will not interfere with any allowed communication or technology.
F29. R6 states “any combination of electro-mechanical items provided by the team that assists their robot in identifying their starting platform” – does this mean there is no communication allowed between the home beacon and the robot(s)?
No, teams are allowed to communicate between their robot and home beacon (per C41), provided it complies will all Disallowed Technology rules and FCC regulations.
F30. Could clarification be provided on some of the following:
Is the 80,000 square meters roving area one long strip, round, square, rectangular? (E7)
See rule E7. Approximately 6 months prior, teams will be provided with the imagery of the Level 1 and Level 2 courses. This will include “the area of interest for the pre-cached sample” as well as areas of interest of Level 2 samples.
F31. Could you clarify R17 and how the pause switch is supposed to work?
Yes. It is intended that the pause switch be a robust switch that, when triggered by an event official, sends a signal to the robot to pause all motion. When triggered again, another signal is sent which tells the robot it may resume motion. Among other things, since it is possible that a single team entry could require multiple pause switches (i.e. for spawn), the pause switch should not be designed such that an event official has to continually hold the button for the duration of the run or for the duration of the pause in order for the robot to remain in that state. We envision, as an example, a garage door opener as a simple potential solution. Teams should plan for these buttons to be robust, easy to use, and easy to hold because the onus is on the teams to ensure the switch works and remains active for the duration of the run.
F32. How will you deal with samples that may roll or move because of wind, being hit by a robot, or being hit by an event official?
We anticipate placing samples such that they will not move because of natural (i.e. wind) forces. However, in any situation where movement of a sample is caused by natural forces or robot interaction, the sample will not be replaced to its original spot and it will ‘play as it lies’. In these cases, it is possible a sample will move closer to the boundary fence than 1 meter or become within 10 meters of another sample. In the event that a sample is hit or moved by an unnatural or non-robot force (i.e. event official) it will be replaced as close as possible to its original spot.
F33. What happens if an official inadvertently triggers an e-stop in the middle of a run?
We feel that the potential of an e-stop being accidentally or unintentionally triggered during a run is very, very small. The scenarios in which this could occur are hard to imagine and nearly impossible to name outcomes for at this time. If this were to happen in the challenge, on-site judges would convene to evaluate the situation and determine an appropriate resolution depending on the exact situation and circumstances. Some examples of potential outcomes we believe would be considered are: restarting the robot in base with the balance of time remaining and field in current status, restarting the entire run from base including removing and replacing any samples collected, or stopping the run and evaluating the team’s performance based on the field as it currently stands. In no case will an inadvertent e-stop cause a disqualification of a team.
F34. Will you be providing additional information on the hard samples?
Yes. All fully registered teams will be provided with the potential rectilinear markings for the hard samples. Only samples with those markings are counted, but as per FAQ F2, we will not intentionally be placing false samples on the course.
F35. Are you planning to fully reveal the challenge location with either the topological data or satellite imagery?
Approximately six months from the event, we will release appropriate topographical data and imagery of the course to aid all competitors in successfully achieving the challenge. It is intended for this information to mimic the information a satellite or previous rover may have collected about the area. It is not our intention to reveal the actual location of the challenge at that time. While it is possible a team may be able to guess a potential location from the information, it will not be confirmed until teams arrive at WPI and are then transported to the event.
F36. Is it possible that the starting platforms will be set up as a “chute” entering the contest area with fencing on either side? Is it possible that the starting platform will be set up pointed directly at a close (< 2 m) snow fence?
Yes, it is possible the starting platforms will be setup with a ‘chute’ leading to an open area of the field. However, the robot will never be started pointed directly at a fence closer than 2 meters.
F37. Can rule E10 be interpreted to mean the hard sample could be 20x20x20cm?
At this time, we will not be providing any additional details or information on the hard samples. However, recall that the challenge is about searching for and identifying samples, so we do not anticipate any special manipulation being required to handle the hard samples versus the easy or intermediate samples.
F38. May I test my robot on-site prior to the event?
Absolutely no testing of robots will be allowed on-site prior to the event. The rules and regulations of the competition site directly prohibit various activities, specifically the operation of any sort of motorized vehicle. Violating any rules of the site would likely disrupt and delay the schedule of the entire event (a probable result of your actions is that the property owners rescind their offer to allow us to host the competition there) and thus doing so would be considered a violation of Section 4 of the Team Agreement.
F39. Will samples be placed inside any structures in the competition area?
No. Samples will not be placed inside any structures (i.e. buildings, trash cans, etc) on the course.
F40. May we use a solar tracking sensor?
There are no rules that prohibit the tracking of celestial bodies like the sun.
F41. What are the eligibility requirements for receiving awards? Is it open to the world this year, or still only primarily US citizens?
All teams are welcome to participate, but only US teams who meet the criteria outlined in the Team Agreement are eligible to win the $1.39M in prize money outlined in Sections 4.2 and 4.3 of the rules.
However, all officially registered teams who meet the deadlines and metrics set forth in the rules and Team Agreements will be eligible for the Technology Achievement Awards described in Section 4.1 of the rules.
F42. What happens to the prize distribution if a demonstration team (not prize eligible) places in the top three for Level 2?
If a team places in the top three of Level 2 but is ineligible for prize money per the team agreement, the money will be distributed to the top three prize eligible teams per the outlined structure in P8. An example is below.
Example:
The top 4 teams score 13, 8, 5, and 4 points respectively. However, the team who scored 10 points is a ‘demonstration’ team and not eligible for the prize. Therefore, the team who scored 8 is considered the top-scoring team for prize distribution and $750,000 is available for distribution per (P8).
F43. R15 does not say what the fans are cooling. Are there any restrictions?
No. This year we do not restrict what fans can cool. However, onus will be on the team to prove any fans on their robot are not doing other actions besides cooling (like the sample collection).
F44. Previous years you suggested that the pause switch be a ‘two-button pause’. Is that suggestion the same for this year?
Yes, a