Competition Rules [Metallic Surface/Buried Object Detection + Minefield Mapping]

Minesweepers is an international competition for humanitarian demining. Each participating team (Max. 10 members) will construct a teleoperated or an autonomous robot that should be able to search for underground and aboveground anti-personnel mines and produce automatically a map of the detected mines. The robot has to able to navigate through rough environment that mimics a real minefield.

1. Minefield

The competition environment will be an open wood area with a size of 20 x 20 m, delimited by 4 GPS coordinates. The competition area will be marked by plastic tape for visualization purposes, there will also be a virtual fence to stop the robots from going outside the competition area. There will be no mines in a stripe of 0.5 m of terrain along the borders of the competition area. Most of the arena will be covered by low grass with a few trees, some steep inclines, ditches and culverts. Some photos of the minefield are shown below.

2. The Mines

Two different kinds of artificial mines are used in this competition:

Buried Mines: These mines are made from metallic cubes, with approximate dimensions of 10x10x10 (LxWxD). These mines are completely buried underground with maximum depth 10 cm. These buried metallic cubes mimic real anti-personnel (AP) blast mines. Real AP blast mines are deliberately designed to be small (typically 6-14 cm in diameter): this makes them cheaper and easier to store, carry and deploy. AP blast mines rely on the effect of explosive blast to damage the victim, and are designed to detonate when the victim steps on them. These mines are often buried in order to camouflage their presence.

Surface Mines: These mines are made from metallic cubes, with approximate dimensions of 10x10x10 (LxWxD) and labeled in black color.  These mines are visible and are located on the surface of the competition area. Any contact with them will be penalized. These black metallic cubes are used to simulate aboveground mines and unexploded ordnance (UXOs). Unexploded Ordnance is a piece of explosive ordnance or ammunition that has failed to function as intended. Although they have failed to function as intended, UXO can sometimes require only the slightest disturbance to detonate. UXOs vary greatly in size from hand grenades the size of an apple to large aircraft bombs.

Some landmines will be organized in a pattern for easier removal and accountability and others will be scattered randomly. Locations of each landmine will be known for the jury committee.

3.  The Robots

Each team must use a teleoperated or an autonomous robot per game. The robot has to be made by team members. Teleoperated robot must be operated remotely from a base station located outside the minefield. Wireless controller based on ZigBee for example would be recommended to communicate the base station with the robot due to the large size of the field. In case of autonomous robots, all the actions of the robot must be completely autonomous without human intervention. Autonomous robot will be rewarded a 40% bonus over teleoperated robots. Careful attention must be paid to the robot locomotion systems as the roughness of the terrain is very high. Both unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs) are allowed as illustrated in the figure below.

Unmanned ground vehicles can be wheeled, legged or hybrid. Wheeled robots include but are not limited to differential drive, tricycle drive, Ackerman steering, synchro drive, omnidirectional drive, Multi-Degree-of-Freedom (MDOF) vehicles, MDOF vehicle with compliant linkage or tracked vehicles. Legged robot can be uniped, biped, tripod, pentapod, quadruped or hexapod robot. Any types of hybrid locomotion can also be used. Examples of hybrid locomotion include a vehicle equipped with tracks for fast locomotion, and legs for more difficult terrain or flippers with self-cleaning tracks or legged vehicles with driving wheels attached to the end of each leg. UAV can be an alternative for the locomotion systems. As shown in above figure, UAVs can be classified into fixed wing, conventional helicopters and multi-rotor helicopters. Fix wind UAVs are naturally stable platforms capable of long flight times and extreme range. However, they are difficult to coordinate with slower ground systems. Conventional helicopters are common collective pitch model helicopters known for their excellent maneuverability and scalability. However they suffer from high level of complexity. Quad-rotors have the ability to hover and there are naturally stable and durable. However, they have limited pay-load. Ball-bots or UAVs that can land on a roving platform are also allowed.

Robot can also be an unmanned aerial vehicle or a quad-rotor. Robot can be actuated using electric, pneumatic or hydraulic actuation system, Diesel/Petrol engine or using solar energy.

4.  Sensors

Each team can select their own set of sensors for localization of mines. Although teams can install cameras on robot or install them on the sides of the field, no camera or sensors is allowed to hangover the competition area.

5.  Mine Detection

When a robot detects a mine, it has to autonomously report this event using a light blinking signal and a warning siren for at least 2 seconds. Teams have to correctly position the alarm device on their robot.

6.  Mine Map

Each deminer robot has to provide map of the detected mines when it's competition time slot finishes. The map represents a 19x19 meter area divided to 100x100 cm squares based on the common reference frame of the arena. The X coordinate of the map is shown by letters A to S and the Y coordinate of the map consists of numbers 1 to 19. So a position (x,y) where the mine is detected has to be reported for example by B2. This map can be simply a text file or text shown on the display of the robot. The sequence of the positions has to be the same as the detected mines. This mine map can be represented graphically or using vector format as shown in Fig. 1 and 2 respectively.

Fig. 1 Graphical Representation of the Mine Map [Black: buried mine, Gray: Surface Mine]

Fig. 2 Vector Representation of the Mine Map

The scoring points will be calculated from the map. If the map is not created automatically by the robot without operator intervention and submitted to the jury committee, the judge won't be able to complete the scoring sheet. There will be two judges: one in the arena and one outside. The judge in the field will take care of observing the touching with the surface mines and control the time and the resets. But the judge outside the arena will calculate the points based on the provided map taking into consideration the observations of the in-field judge in terms of number of touches and time. If no map is provided (in raw or graphic format as shown below), no points will be obtained.

7.  Procedure

Each robot starts the game from one of the corners of the competition arena such as A1 (Fig. 1). Team members will bring the teleoperated or the autonomous robot to this location. Then the robot has to search the field to find buried mines or the mines scattered on surface. When the robot detects any kinds of mine it should register the location of the mine in the map and produce a light signal and siren and also report the mine location to update the mine map. All the detected mines will be removed from the field before a new team enters the arena.

Robot has to able to navigate through rough environment of the minefield and avoid obstacles. Robots must avoid surface mines else the team will be penalized.

During competitions only one of team members can attend the field. He/she can request a “Reset Time” which means he/she can stop the game and take out robot for repair or adjustment. The time spends for this repair will be included within the competition time and there would be a penalty for each resent time. The competition time allowed for each team is 20 minutes including the reset time. Jury committee will calculate the team's score and prepare the field for the next team during another 10 minutes.

  • The competition will end with one of the following conditions:

  • The dedicated time finishes,

  • Team dismiss the game,

  • Any cheating happens,

  • Robot touches a surface mine.

8.  ROS

ROS

The main requirement to participate in this category is to build the software system of the robot using Robot Operating System (ROS). ROS is an open source platform that provides a set of software libraries and tools to help you build robot applications. It provides hardware abstraction, device drivers, libraries, visualizers, message-passing, package management, and more. Building the robot based on ROS allows you to make use of interesting software module and the high fidelity simulator developed by the participants of the different editions of Humanitarian Robotics and Automation Technology Challenge (HRATC) organized by IEEE Robotics & Automation Society – Special Interest Group on Humanitarian Technology (RAS–SIGHT). Minesweepers is also organised under the auspices of IEEE RAS SIGHT. ROS-based robots can participate in any of other four categories of the competition. Team with ROS-based robots can be awarded the prize of any of the other four categories plus a special award for this special category.

Hardware: You can build your own robot and make it ROS-enabled or you can use low-cost ROS-enabled robotic starter kits.

Software: You can write your own ROS code or you can download and use ROS Open Source Modules for Humanitarian Demining: SIGHT - Special Interest Group on Humanitarian Technology Open Repositories

 

ROS Tutorials:

  • What is ROS?

  • ROS (Robot Operating System) is a flexible framework for writing robot software provides libraries and tools to help software developers create robot applications. It provides hardware abstraction, device drivers, libraries, visualizers, message-passing, package management, and more. ROS is completely open source (BSD) and free for others to use, change and commercialize. It aims to simplify the task of creating complex and robust robot behavior across a wide variety of robotic platforms and enable software developers to build more capable robot applications quickly and easily on a common platform. 

 

9.  Multi-Robot Systems (MRS)

Multi-robot systems (MRS) are a group of robots that are designed aiming to perform some collective behavior. By this collective behavior, some goals that are impossible for a single robot to achieve become feasible and attainable. There are various foreseen benefits of MRS compared to single robot systems. These benefits include, but are not limited to the following [Khamis 2014]:

  • Resolving task complexity: some tasks may be quite complex for a single robot to do or even it might be impossible. This complexity may be also due to the distributed nature of the tasks and/or the diversity of the tasks in terms of different requirements. Examples of these tasks include reconnaissance, surveillance, search and rescue.
  • Increasing the performance: performance measures are application-dependent. However, and as an example, task completion time can be dramatically decreased if many robots cooperate to do the tasks in parallel. Spatial and/or temporal area/object coverage can be improved using multiple robots. Moreover, in some applications, these robots can cooperate to establish ad hoc communication relay network to improve radio coverage.
  • Increasing reliability: increasing the system reliability through redundancy because having only one robot may work as a bottleneck for the whole system especially in critical times. But when having multiple robots doing a task and one fails, others could still do the job.
  • Simplicity in design: having small, simple robots will be easier and cheaper to implement than having only single powerful complex robot.

MRS

MRS can play a crucial role in humanitarian demining. According to the Standard Operating Procedures (SOPs) for humanitarian demining, human deminers use metal detectors to identify targets, which are then flagged for subsequent digging by a supervisor. The objective of this category is to mimic the conventional mag-and-flag approach or SOP using multiple unmanned teleoperated and autonomous vehicles. The arena of the competition is shown in the following figure. Teleoperated vehicles play the role of human deminers while an autonomous vehicle is used to mimic the supervisor/team leader’s role. The team leader has to be equipped with a gripper or a marking mechanism to mark the location of the landmine detected by the deminers. More than one deminer can be integrated into the team but only one supervisor or team leader is allowed.

If two deminers are used, these unmanned vehicles “A” and “B” are assigned to each lane as shown in the figure. Vehicle “B” starts to work after vehicle “A”. If a deminer detected a surface-laid or a buried mine in its assigned lane, it has to produce a warning siren for at least 2 seconds and inform the team leader about the position of the detected mine. All the deminers have to stop and wait while the autonomous vehicle (the team leader) comes forward marks the detected mine with a red mark or flag and the scanning procedure continues until all the arena is scanned.

General rules for scoring are as following. But the exact scores will be decided during the competition.

  • 10 Positive score for detecting every buried metallic mine by the deminers,
  • 5 Positive score for detecting every surface metallic mine by the deminers,
  • 30 Positive score for complete surf of field if 80% of mines are detected correctly by the robot team,
  • 5 Negative score for wrong detection by the deminers,
  • 10 Negative score for passing over a buried mine without detecting it by the deminers.
  • 5 Negative score for touching a surface mine by the deminers.
  • 3 Negative score failure in producing a light signal and/or a siren by the deminers for a detected mine.
  • 5 Negative score per deminer for failure of pausing the movement of the deminer after detecting the landmine and failure in waiting the supervisor.
  • 10 Negative score for failure of correct marking by the supervisor.
    • 2 Negative score for every minute reset time.

 

10.  Scoring sheet

 

Action

Count

Score/Unit

Subtotal

Arena Score: The following score will be based on the performance of the robot in the competition arena and will be observed and calculated by the in-field judge.

Detected Surface Mines

 

5

 

Detected  Underground Mines

 

10

 

Completely Scan the field  and 80% of Mines Detected (Systematic Motion)

Yes

No

30

0

 

Wrong Detection of a Mine

 

-5

 

Passover Buried Mine without Detection

 

-10

 

Touching Surface Mine

 

-5

 

No light signal and/or a siren

Yes

No

-3

0

 

Reset Time (-2/1min)

 

-2

 

Mine Map Score: A mine map (in vector or graphical format) must be created automatically by the robot system and presented to the out-field jury committee by the team representative to calculate the following scores:

Mine Map True Positive for Surface Mines (a minefield cell contaminated by surface mine is labeled in the map as a cell with surface mine)

 

5

 

Mine Map True Positive for Buried Mines (a minefield cell contaminated by buried mine is labeled in the map as a cell with buried mine)

 

10

 

Mine Map False Negative for Surface or Buried Mines (a minefield cell contaminated by surface or buried mine is labeled in the map as a clean cell)

 

-5

 

Mine Map False Positive for Surface or Buried Mines (a clean cell in the minefield is labeled as contaminated by surface or buried mine is the mine map)

 

-5

 

Total Score:

 

      Autonomous Robot

☑   YES (Multiply by 1.2)

◻   NO

      Running Using ROS

☑   YES (Multiply by 1.2)

◻   NO

    Multi Robot System

☑   YES (Multiply by 1.2)

◻   NO

Final Score: