RoboCupRescue Robot League

From Robocup

Jump to: navigation, search

SCORES!!!!!!!

CONGRAULATIONS TO ALL TEAMS. JOB WELL DONE.

CHAMPIONSHIP: MIXED INITIATIVE MISSIONS
Teams with the highest cumulative scores from 7-10 missions receive 1st, 2nd, 3rd place awards
1st Place: INDEPENDENT, KING MONGKUT'S INST. OF TECH. BANGKOK, THAILAND
2nd Place: PELICAN UNITED, CHIBA INSTITUTE OF TECH. & TOHOKU UNIV., JAPAN
3rd Place: CEO MISSION, UNIV. OF THE THAI CHAMBER OF COMMERCE, THAILAND


BEST-IN-CLASS: MOBILITY
Robots that found the most Red Arena victims throughout and scored the most points in mobility missions
1st Place: PELICAN UNITED, CHIBA INSTITUTE OF TECHNOLOGY, JAPAN
2nd Place: SHINOBI, THE UNIVERSITY OF ELECTRO-COMMUNICATIONS SGI, JAPAN
3rd Place: INDEPENDENT, KING MONGKUT'S INST. OF TECH. BANGKOK, THAILAND

BEST-IN-CLASS: AUTONOMY:
Robots that found the most Yellow Arena victims throughout and scored the most points in mapping missions
1st Place: RESKO, UNIVERSITAT KOBLENZ UND LANDAU, GERMANY
2nd Place: JACOBS RESCUE ROBOT, JACOBS UNIVERSITY BREMEN, GERMANY
3rd Place: RFC UPPSALA, UPPSALA UNIVERSITY, SWEDEN

Good Maps

Jacobs2.jpg Pelican2.jpg Resko.jpg Uppsala.jpg

Preliminary Schedule

Sunday, July 1: Setup

1200 Campus Recreation Center (CRC) is open to teams
Teams setup on paddock
2200 Venue closed


Monday, July 2: Setup and Practice

0700 Teams may practice in the arenas
2200 Venue closed

Tuesday, July 3: Preliminaries

0700 Teams may practice in the arenas

1000 Opening Ceremony in the Campus Recreation Center

1200 1st mission (preliminaries with up to 24 teams)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 20 minutes (12 teams in each arena @ 2 per hour = 6 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1800 Teams may practice in the arenas

2200 Venue closed

Wednesday, July 4: Preliminaries

0900 2nd mission (preliminaries with up to 24 teams)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 20 minutes (12 teams in each arena @ 2 per hour = 6 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1500 3rd mission (preliminaries with up to 24 teams)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 20 minutes (12 teams in each arena @ 2 per hour = 6 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

2200 Venue closed


Thursday, July 5: Preliminaries and Combined Rescue League Workshop


0700 Teams may practice in the arenas

0900 4th mission (preliminaries with up to 24 teams)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 20 minutes (12 teams in each arena @ 2 per hour = 6 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1600 SSRMED-2007: Combined rescue leagues workshop

Hosted near rescue robot league arenas for robot demos, presentations, and discussions across all rescue leagues

1800 Teams may practice in arena
Arenas get reconfigured for Semifinals but stay separate and concurrent.

2200 Venue closed


Friday, July 6: Semifinals


0700 Teams may practice in the arenas

1000 5th mission (semifinals with 10-12 teams based on scoring gap)

  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1400 6th mission (semifinals with 10-12 teams based on scoring gap)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 20 minutes (5-6 teams in each arena @ 2 per hour = 3 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

2200 Venue closed


Saturday, July 7: Semifinals


0700 Teams may practice in the arenas

0900 Venue open to visitors

1000 7th mission (semifinals with 10-12 teams based on scoring gap)

  • Arena:Qualifying size (arena A & B working concurrently)
  • Duration: 25 minutes (5-6 teams in each arena @ 2 per hour = 3 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1400 8th mission (semifinals with 10-12 teams based on scoring gap)

  • Arena: Qualifying size (arena A & B working concurrently)
  • Duration: 25 minutes (5-6 teams in each arena @ 2 per hour = 3 hours)
  • Start times: A = 00/30 minutes, B = 10/40 minutes past the hour

1800 Venue closed to visitors, teams may practice in arenas

Arenas get combined to form Championship size arena.

2200 Venue closed


Sunday, July 8: Finals in Championship Size Arena (two arenas combined)


0700 Teams may practice in the arenas

1000 9th mission (finals with 5-6 teams based on scoring gap)

  • Arena: Championship size (A + B)
  • Duration: 25 minutes (5-6 teams in the arena = 3 hours)
  • Start times: A = 00/30 minutes past the hour

1300 10th mission (finals with 5-6 teams based on scoring gap)

  • Arena: Championship size (arena A + B combined)
  • Duration: 25 minutes (5-6 teams in the arena = 3 hours)
  • Start times: A = 00/30 minutes past the hour

1700 Awards ceremony in the CRC

2000 Venue closed



RoboCupRescue Robot League

Competition Overview

The goal of the urban search and rescue (US&R) robot competitions is to increase awareness of the challenges involved in search and rescue applications, provide objective evaluation of robotic implementations in representative environments, and promote collaboration between researchers. It requires robots to demonstrate their capabilities in mobility, sensory perception, planning, mapping, and practical operator interfaces, while searching for simulated victims in unstructured environments. As robot teams begin demonstrating repeated successes against the obstacles posed in the arenas, the level of difficulty will be increased accordingly so that the arenas provide a stepping-stone from the laboratory to the real world. Meanwhile, the yearly competitions will provide direct comparison of robotic approaches, objective performance evaluation, and a public proving ground for field-able robotic systems that will ultimately be used to save lives.


Competition Vision

When disaster happens, minimize risk to search and rescue personnel, while increasing victim survival rates, by fielding teams of collaborative robots which can:


  • Autonomously negotiate compromised and collapsed structures


  • Find victims and ascertain their conditions


  • Produce practical maps of their locations


  • Identify hazards


  • Emplace sensors (acoustic, thermal, hazmat, seismic, etc,…)


  • Provide structural shoring


…allowing human rescuers to quickly locate and extract victims.

It is ideal for the robots to be capable of all the tasks outlined in the vision, with the first three directly encouraged in the current performance metric. The remaining tasks will be emphasized in future versions of the performance metric.


Search Scenario

A building has partially collapsed due to earthquake.

The Incident Commander in charge of rescue operations at the disaster site, fearing secondary collapses from aftershocks, has asked for teams of robots to immediately search the interior of the building for victims.

The mission for the robots and their operators is to find victims, determine their situation, state, and location, and then report back their findings in a map of the building and a victim data sheet.

The section near the building entrance appears relatively intact while the interior of the structure exhibits increasing degrees of collapse. Robots must negotiate the lightly damaged areas prior to encountering more challenging obstacles and rubble.
The robots are considered expendable in case of difficulty.


League Objectives for 2007

Discussion

One major goal of this competition is to encourage development of capable robots with intuitive operator interfaces that can negotiate increasingly complex and difficult indoor and outdoor urban environments. Over the past few years, teams in this league have moved quickly to develop and incorporate best-in-class solutions demonstrating innovative mechanisms, emerging sensors, promising algorithms, and intuitive operator interfaces that have proven increasingly effective in constantly challenging arenas. The easy lessons regarding the ineffectiveness of simple video streams and remote control vehicles for this application domain have been learned. While ongoing issues regarding communications, especially indoors and down range, will continue to pose challenges to effective remote situational awareness in teleoperative systems. It is apparent that some level of autonomy is necessary to improve the overall performance of these robots while easing the burden on operators working for long periods in critical situations.


This league of teams, this competition, is where integration of these capabilities should happen. Teams and robots with complementary capabilities must collaborate to produce effective systems. The competition has established best-in-class missions within the overall RoboCupRescue events to highlight successful approaches with awards for locomotion and autonomy so that other teams know where to look for other pieces of the robot puzzle. Now is a good time to review last year's awardee TDPs posted on the website http://www.isd.mel.nist.gov/projects/USAR/2006/team_info_bremen_2006.htm and get in touch with their team leaders to collaborate around the upcoming competition. Your teams will benefit. But more importantly, we'll move much closer to the ultimate goal of fielding capable robots for emergency response.


Toward that end, the arenas this year are made up almost exclusively of emerging standard test methods for response robots that will be used by emergency responders to compare/quantify performance for robots available for purchase. Specifically, these standard test methods, which this league has been instrumental in helping to develop, will be published by ASTM International and used by U.S. FEMA US&R teams to evaluate robot performance. Some of the test methods that make up the arenas include visual acuity, steps, gaps, ramps, stairs, stepfields for mobility/endurance, and random mazes to test human-robot interactions and mapping. Directed perception test methods are used to conceal some simulated victims. Others test methods under development but not in the arenas include radio communications, grasping dexterity, and others that will likely appear in future years of this competition.


As always, robots may be remotely teleoperated, although certain assistive autonomous capabilities should improve overall performance. For example, navigating this year's maze with rolling/pitching floors could benefit from wall following, centering, and other automatic behaviors performed by the robot to ease the burden on the operator. Sensor generated maps of the maze could reduce re-tracing already searched areas and back tracking. Background processes to detect victim signs of life could allow operators to focus on negotiating difficult terrain and planning their mission routes. As a real world example of where computational capabilities improve robot performance without necessarily reducing the operator's role, think of a robot operator with a serial-link manipulator and a coordinated control system and a joystick vs. an operator with basic joint level switches and dials. Then understand that most response robots currently deployed with manipulators use joint level controls for remote teleoperative tasks – there is plenty of room for improvement. This league of teams can do better. We need to work together as a league to incorporate the advanced technologies we work with every day, while keeping an eye on how best to use them in this application domain. This year's arena is a full step in that direction. The winners of this year's competition, with some hardening and refinement, will be on the edge of successfully answering responders' needs. Your efforts toward that end are much appreciated, and your successes will be advertised to responders around the world so that they know the kinds of capabilities they can start expecting, even demanding, from robots in the field.


Mission Strategies


To emphasize efficient team deployment strategies, setup times will be included in overall mission durations. Preliminary/semi-final round missions will be 15 minutes long, which includes setup time in the operator station and placement of the robot at the start point (with start times every 20 minutes). Final round missions will be 25 minutes long (with start times every 30 minutes). Teams will be expected to queue in advance of their scheduled start time. Teams that can deploy quickly should have more time for searching within the arenas.


All robots will start their missions within the Yellow arena to increase the size of the arena to be searched. However, only robots able to autonomously negotiate the Yellow arena and detect a victim's signs of life can score Yellow arena victims. Autonomous in this case means that after setting up the robot at the start point, the operator can simply press "start" and watch his/her operator interface – hands off. The operator may control other robots during this time (probably searching in front of the autonomous robot to avoid blocking a hallway). The operator may take over teleoperation of the robot at any time, but all teleoperated robots must find their way through the Yellow arena maze to the Orange and/or Red arenas to find and score victims.


To find and score Yellow arena victims, the robots must autonomously stop and display to the operator and judge the particular signs of life that it has detected. The operator and judge will look at the signs of life being presented by the robot. If the judge can identify the victim's signs of life based on the view/information provided at the time the robot called for the operator's attention, the victim will be mapped and the robot allowed to resume searching (a single key stroke). If the judge cannot identify the signs of life from the view/information provided, the robot should NOT map the location and resume searching (a different key stroke). Watching as the robot approaches doesn't count unless the robot saves the view/information for display when it calls for the operator. The process for verifying the victim information will take 30 seconds, so there are a limited number of mistakes that can be made before time runs out. At least two co-located signs of life should be detected by the robot before stopping and notifying the operator to avoid false positive identifications.

Remotely teleoperated robots will need to negotiate and map the Yellow arena on their way to finding the Orange and Red arena box stacks and the victims inside. Red arena victims will be placed in box stacks searchable from the main floor full-cubic (red) stepfields and the elevated floors accessible via the stairs or the steep ramp.


New GeoTIFF map formats

New GeoTIFF map format to submit after missions
To move our league's mapping capabilities one step closer to providing emergency responders with useful maps of building interiors, and tie our league closer still to the Virtual Robot Competition, we'll be asking teams to submit their maps from each mission in a standard geographically-referenced tagged image file format, or GeoTIFF format.

This will allow us to better compare maps to the arena ground truth, and to compare different maps to each other. We'll use the location of the victims in the map as control points for comparison as always, and will discuss on-site the possibility of adding additional hazmat labels in the environment to locate particular walls in the map, thus increasing the overall number of control points on any given map. As always, we're pushing hard to move the league forward faster to solve emergency responders' critical issues.

This league is a collaborative development effort first and foremost, and so several teams have stepped up to provide the tools necessary to generate GeoTIFF files. See the presentation below, generated by Andreas Birk at Jacobs University, Germany at the 2007 German Open, which outlines the approach and presents the key issues for discussion. In Atlanta, we'll be rewarding teams that can provide GeoTIFF formatted files with full credit on their map quality scores (10 pts of 50 for each victim as always) and better comparison to ground truth for map accuracy scores (also 10 pts of 50 for each victim as always). So there is incentive for every team to incorporate this new capability. Responders know about GeoTIFFs for wide area operations and planning. Well be providing interior maps that can be incorporated into existing geographic information systems (GIS) in use by emergency responders.


Rules at a Glance

New This Year

  • Victim placements will be known to the operators and audience prior to missions, and changed each round to ensure complete arena coverage over multiple missions.
  • Resets allow fixing/replacing the robot at the start point but loss of accumulated victims, maps, and time.
  • GeoTIFF map formats will be used to allow comparison of maps to ground truth arena configurations.
  • Best-In-Class awards for autonomy and mobility will be given to robots that find the most victims in the Yellow and Red arenas respectively over all missions.


Arena Features: Yellow, Orange, Red

  • Random mazes with non-flat flooring
  • Stepfield pallets (Orange: half-cubic, Red: full-cubic)
  • Stairs (40°, 20cm riser, 25cm tread depth)
  • Ramp (45° to test torque and center of gravity)
  • Confined spaces (ceiling blocks under elevated floors)
  • Visual acuity (tumbling E eye charts, hazmat labels)
  • Directed perception boxes with victims/targets inside


Simulated Victims: 4 per arena, 12 total

  • The chair will place victims in two high and two low boxes per arena, in different locations each round.
  • Signs of life: form, heat, motion, sound, and/or CO2
  • “Trapped” are in boxes open on top
  • “Void” are in boxes open to side
  • “Entombed” are in boxes with view holes
  • Tumbling E’s and/or hazmat labels are victim tags

Misssions

  • Teams queue at paddock entry prior to scheduled start.
  • 15/20/25 minute missions include robot placement at the start point and operator station setup. Each team is responsible for making sure victims are functional (heat, batteries, tags) prior to their mission start.
  • Teams are allowed one operator during missions.
  • Start points will be in the Yellow arena with all robots facing the same direction (“north” on your map).
  • Yellow arena victims can be scored only by robots with autonomous navigation and victim identification. Operators may take over control at any time to move into the Orange and Red arenas but must return to the start point to resume autonomous searches.
  • Teleoperative robots can only score Orange or Red arena victims, which are placed on both sides of the Yellow arena to encourage complete mapping.
  • Resets allow fixing/replacing the robot at the start point but loss of accumulated victims, maps, and time.
  • Bumping penalties are assessed if the administrator must replace/fix arena elements prior to next mission.
  • GeoTiff map formats get full scores for map quality and will be compared to ground truth for accuracy.
  • Highest cumulative scores from 7-10 missions will be awarded 1st, 2nd, 3rd place awards.
  • Best-In-Class awards will be given to individual robots that do the following during all missions:
  • Autonomy: Find the most Yellow arena victims
  • Mobility: Find the most Red arena victims

Scoring Metric

The competition rules and scoring metric both focus on the basic US&R tasks of identifying live victims, determining the victim's condition, providing an accurate victim location, and enabling victim recovery, all without causing damage to the environment. The teams compete in several missions lasting up to twenty minutes with the winner achieving the highest cumulative score from all missions. The performance metric used for scoring encourages identification of detailed victim information through multiple sensors along with generation of easily understandable and accurate maps of the environment. It also encourages teams to minimize the number of operators, which may be achieved through using better operator interfaces and/or autonomous behaviors that allow effective control of multiple robots. The scoring metric also discourages uncontrolled bumping behaviors that may cause secondary collapses or further injure victims. Since all robots compete within the same arena, no weighting is necessary to account for arena diffulty.



Score.jpg


Video and Data Collection


We will try to collect quad-screen video this year for all semi-final and final missions. This video format captures concurrent streams of key video to help understand what is working in your system, and challenges left to solve. The four video streams are the robot within the arena (upper left), robot position tracking within the maze (upper right), operator interface interactions (lower left), and operator interface video (lower right).
Display.jpg



Arena Elements


Random maze with non-flat floors


This year's Yellow, Orange, and Red arenas form a continuous, randomly generated maze. The maze consists of 1.2m wide hallways, defined by1.2m square walls and corners, with horizontal crossbars over top for support. Robots will need to pass under these occasional crossbars. Robots will also need to negotiate a variety of non-flat flooring with increasing complexity as they progress through the arenas (no more assumed flat flooring).

Simulated victims are distributed equally across the three arenas. Robots capable of autonomously negotiating and finding victims in the Yellow arena have access to 1/3 more victims to find and score before moving on to be remotely teleoperated through the Orange and Red arenas. We hope that is incentive enough to work on or team with organizations skilled in the basic autonomous behaviors that can improve performance.

Yellow Arena


All robots will begin their mission in the Yellow arena maze to directly compare navigation, victim identification, and mapping capabilities in autonomous robots along with remote situational awareness and mapping capabilities in remotely teleoperated robots. This year's yellow arena flooring will include 10-degree pitch/roll ramps filling the hallways between the walls (see figure below). Preliminary round missions will have mostly flat flooring (covered with paper) with intermittent pitch/roll ramps. Final round missions will increase the complexity to be mostly pitch/roll ramps. Within the maze there will be a variety of sensory obstacles for:

  • ultrasonic range sensors (absorptive ceiling tiles and reflective corner angles),


  • laser range sensors (absorptive dark felt, reflective mirrors, and transparent plexiglass),


  • victim identification sensors (false sources of heat, motion, and/or sound).


Yellow1.jpg Yellow2.jpg Yellow3.jpg
Figure 1:Yellow arena maze with non-flat flooring consists of A) maze walls, B) intermittent pitch/roll ramps and open face/top box stacks containing simulated victims, and C) continuous pitch and roll ramps (in final rounds)



Orange Arena


The Orange arena maze will include continuous 15-degree pitch/roll ramps and half-cubic (orange) stepfields (see below).

Orange1.jpg Orange2.jpg Orange3.jpg
Figure 2: Orange arena maze with more complex non-flat flooring contains A) continuous pitch/roll ramps, B, C) half-cubic stepfields and box stacks with simulated victims that require directed perception and variable illumination to see inside.




Red Arena


The Red arena maze will include continuous full-cubic (red) stepfields as flooring. There will also be elevated floors accessible by stairs (40 degree incline, 20cm step heights, 5 steps total) and a steep ramp (35-45 degree incline with carpet for traction) to challenge power and center of gravity issues. Each elevated floor section will be separated by a 20cm step/pipe combination to challenge robots reliant on the sharpness of step edges for traction and reward robots that can change their shape to reliably surmount curbs of any condition. Confined spaces under the elevated floors will have stalactites hanging down and may be placed over ramps or stepfields.

Red1.jpg Red2.jpg Red3.jpg
Figure 3: Red arena maze includes A,B) full-cubic (red) stepfield “hills,” “flats” and “diagonals,” along with C) elevated floor sections accessible via stairs or a steep ramp, with stalactites hanging underneath to form confined spaces.

Red4.jpg Red5.jpg Red6.jpg
Figure 4: Other Red arena elements include A) stairs with 40 degree incline and 20cm step heights, a 35-45 degree ramp with carpet to access the elevated floors (not shown), B) 20cm step/pipe combinations to minimize corner traction divide elevated floor sections, and C) confined spaces under elevated floors (also used with ramp flooring).

Box stacks containing simulated victims


All simulated victims will be placed in cardboard boxes with at least three signs of life: always human form and heat, with sound, motion, and/or CO2. Multi-level box stacks provide searchable voids that may contain victims at various heights (45cm/18in boxes, up to three levels high). Up to six box stacks will be placed across all three arenas, roughly two per arena, and colored appropriately for arena identification. Simulated victims will be distributed equally across all three arenas: three or four victims per arena placed on assorted sides/levels of the box stacks within that arena.

"Void" victims will be contained in box stacks with open faces and tops making all the victim's signs of life rather obvious: human form (baby or arm), heating pad (mounted to back of box), motion (waving or twitching), tape recording (yelling or crying), CO2 (bottles/cartridges opened and placed in container).

"Entombed" victims will be contained in boxes with 15cm holes requiring sensor placement directly in front of the holes to identify sensory targets inside (eye charts and/or hazmat labels inside count as "form" points). Variable illumination near the camera is important and absolutely essential for fieldable response robots.

Box1.jpg Box2.jpg Box3.jpg

Figure 5: Box stacks with 1-3 levels contain simulated victims. A) Open boxes contain “void” victims that are viewable from face and top. B) Boxes with holes contain more difficult “entombed” victims that are also viewable from the face and top. C) All boxes contain eye charts (tumbling E’s) directly in line with the holes and hazardous materials labels off to the side which can be identified and scored in place of human form points along with the other victim signs of life.

Void1.jpg Void2.jpg Void3.jpg

Figure 6: ”Void” victims found within the Yellow arena for autonomous robot identification and some harerd to reach Red arena locations. A) Shown with flat flooring, and B) with pitch/roll ramps in finals. C) Eye chart, hazmat label, and all victim signs of life can be identified and scored.


Entombed1.jpg Entombed2.jpg Entombed3.jpg
Figure 7: “Entombed” victims found within the Orange arena and easier sections of the Red arena require more dexterous placement of cameras and sensors to look into holes. A) Shown among pitch/roll ramps. B) Shown among half-cubic (orange) stepfields. C) Looking through the hole to identify the eye chart, hazmat label, and all victim signs of life. Note that variable illumination near the camera is very helpful and an absolute requirement for response robot applications.

Other League Initiatives

German Open highlights

Successful German Open highlights new league initiatives for Atlanta
The RoboCupRescue competition at the 2007 German Open provided an intense development session to move the league forward in vital areas: generation of new GeoTIFF map formats to compare mission results with arena ground truth.; capture/dissemination of sensor data sets from missions within the arenas; and co-location with RoboCup@Home League to support increased collaboration among teams. See the pdf below for an overview of the competition:

PDF coming soon!

Sensor data sets captured within arenas

The pointer below provides access to sensor data sets that were collected in the RoboCupRescue Robot League competition arena during the RoboCup German Open at Hannover Fair in April 2007. It consists of recorded streams of raw sensor data and motor commands of a so-called Jacobs rugbot ("rugged robot"). The main design purpose of this robot is for research in rescue robotics [3]. The data and the related software can be freely used for academic purposes. Please use references to the Jacobs robots [1][2] when you generate your own work based on this data.
Please contact Robotics at Jacobs University (http://robotics.iu-bremen.de) if you want to use the data or the related software for non-academic purposes.

http://robotics.iu-bremen.de/datasets/RoboCupGermanOpen2007/


Co-Located with RoboCup @Home League


Starting this year, the Rescue Robot League will be co-located with @Home League to form a continuum of challenges for autonomous robots (full autonomy is a requirement in the @Home League). Essentially, the @Home arena will form the "intact" living space of a small house. There will be two doors to exit that space and enter directly into the Rescue Robot League's Yellow arena maze with rolling/pitching floors. Autonomous robots with reasonably mobile chassis/wheel choices will be able to handle the transitions/thresholds (see figure below). The doors will be open this year and probably next (@Home may emphasize door opening tasks starting this year). Typical interior doors will also be used to define entrance/exit points within the Rescue Robot League arenas as well. These doors will be left open this year, but future years will require door opening skills to access certain rooms with bonus victims to find.
Bot1.jpg Bot2.jpg Bot3.jpg
Figure 8: Example autonomous robots with basic mobility that can handle the pitch/roll ramp in the A) Yellow arena and some of the Orange arena, B) Yellow arena and most of the Orange arena including the half-cubic stepfields, and C) all three arenas including the full-cubic stepfields in the Red arena (and maybe even the stairs). See http://www.volksbot.de/ for purchasable kits. These robots will be demonstrated at the German Open.


We hope that your preparations are progressing well. As mentioned previously, the winners this year will be on the edge of functionally effective for emergency response applications. With some hardening and other design modifications (like less duct tape!), you'll likely see these robots in the hands of emergency responders in practice deployment scenarios. Which is the necessary next step toward deployment in the field.



Good luck!



Other Infomation

Important Dates

January 31, 2007


Submit a Team Participation Form http://www.isd.mel.nist.gov/projects/USAR/2007/Champ2007_(Country_TeamName)_TPF.doc to rescue.robot.league@nist.gov


February 15, 2007


Submit a Team Description Paper http://www.isd.mel.nist.gov/projects/USAR/2007/RCR2007_(Country_TeamName)_TDP.doc to rescue.robot.league@nist.gov


Submit a Request for travel support http://www.isd.mel.nist.gov/projects/USAR/2007/Champ2007_(Country_TeamName)_TSF.doc to rescue.robot.league@nist.gov

March 19, 2007

Announcement of Qualified Teams

http://www.isd.mel.nist.gov/projects/USAR/2007/qualified_teams.htm



Travel Support Update


According to the RoboCup 2007 General Chair, travel support for teams is not available at this time. See note below:


"While registration fees are slightly higher than last year, we are working to reduce expenses for teams traveling to RoboCup 2007 by offering inexpensive housing (as low as $35 per person per night). The net cost for a team is about the same or lower than last year.


We are starting to see a lot of requests for travel support, and I regret to say that at present we do not have such funding (we are still working on it though)."


Qualification Process for the World Championship Competition

All RoboCupRescue Robot League teams should use nearby regional open competitions to:

1. practice your deployment strategies
2. familiarizeyourselves with the arenas and rules
3. demonstrate your capabilities to the Technical Committee

Teams that demonstrate competent and reliable systems at the regional open competition, along with a descriptive TDP, will always be favored for inclusion in the World Championships.

For example. this year's RoboCupRescue Robot League regional open competitions include:

- Japan Open
- German Open
- Iran Open
- Dutch Open

We are actively trying (but need help from locals) to establish regional open competitions in other areas too.

Unfortunately, due to the close scheduling of the regional opens and the World Championship each year, the qualification process can't be completely sequential. So the Technical Committee qualifies an initial set of teams for the World Championship based on your team's submitted TDP, and experience in previous regional opens. The Technical Committee usually keeps a few slots open to include a final few capable teams which emerge from the regional open competitions in the current year as well. When there is no regional open competition in the region of the World Championship, the technical committee typically qualifies proportionally more teams from the local region, while encouraging them to try to participate in other regional open competitions as well. So the best route to participate in a World Championship is through a regional open competition, and we encourage all new teams to start there. You'll find the lessons are well worth it.

Generally each year the Technical Committee looks at the following criteria to qualify teams for the World Championship:

1. Your TDP describes improvements to your robot based on lessons learned from successfully competing in a semi-final round in a previous World Championship, or

2. Your TDP describes improvements to your robot based on lessons learned from winning a previous regional open or best-in-class competition, or

3. Your TDP describes a particularly interesting or innovative approach that the Technical Committee considers likely to perform well at the competition even without previous experience at a regional open. This is especially possible if you can demonstrate your capabilities convincingly within representative environments through:

a) video of the robot performing any or all of the requisite capabilities:

- advanced mobility (traversing random stepfields or confined space cubes)
- navigation (wall following, centering between obstacles or constrictions)
- localization and mapping (2D/3D maps, SLAM on non-flat-flooring i.e. pitch/roll ramps, low-profile stepfields)
- directed perception (visual acuity for near/far/dark/light, sensor probing into voids, sometimes with reaching)
- victim identification (fusion of the various sensory signals to improve confidence and reduce errors)
- autonomy (assistive features, bounded intervals, or fully autonomous performance of any or all of the above)
- effective operator interfaces

b) performing any or all of the above in an upcoming regional open competition

Some of the qualifications may be granted by the Technical Committee to include particular countries, technologies that the league should be investigating, or to support other league outreach efforts.




Information about the league is available on the NIST site

http://www.isd.mel.nist.gov/projects/USAR/2007/index.htm --Virts 14:21, 4 April 2007 (EDT)

Personal tools