1. STC Robot 2 Majd Srour, Anis Abboud Under the supervision of: Yotam Elor and Prof. Alfred Bruckstein Optimally Covering an Unknown Environment with Ant-like STC algorithm

2. STC Robot Motivation  Recently, it was shown that a single ant robot (modeled as FSM) can simulate the execution of any arbitrary of Turing machine.  Single ant robot, using pheromones can execute complex single-robot algorithms.  In this project, we used an algorithm that leaves Ant-like pheromones on a surface in order to solve the covering problem.

3. STC Robot The Environment  In order to leave pheromones-like traces, we used a phosphorescent glowing floor, that can be charged by light.  The robot charges the board using an UV LEDs which are installed on the back of the robot.

4. STC Robot Task Description Given a phosphorescent glowing floor like this: The robot divides the floor virtually into sub-cells of its size, and each 4 adjacent sub-cells are considered one cell. The Robot’s task is to cover the floor, passing exactly once over each sub-cell and to return to the same cell where it started.

5. STC Robot Last Project vs This project  On-line STC  Board is saved in the computer  The computer “remembers” which cells are visited.  Normal light conditions  Floor is physically divided into cells  Ant-like STC  Board is not saved in the computer  The robot figures out which cells are visited based on traces it left.  Darkness  Floor is virtually divided into cells

6. STC Robot The Robot  PIC-18 board with a Wireless module.  Two main wheels, and two auxiliary wheels.  These enable it to perform an on-place 90 degree rotation.  Wireless web camera directed up.  A curved spherical mirror, directed down.  4 UV (Ultraviolet) LEDs  Enable the robot to leave green tracks (dots) on the phosphorescent floor. The Robot is built from Lego.

7. STC Robot Image Transformation (1) Because the camera looks up to a spherical mirror, the image captured from the Robot’s camera is distorted, so we apply the same transformation as the previous project: But in the darkness.

8. STC Robot Image Transformation (2)  Problem:  Visited cells are green, obstacles are red.  The transformation produces a grayscale image. We can’t distinguish between obstacles and visited cells  Solution:  Filtering red light, and transforming the image  Identifying visited sub-cells.  Filtering green light, and transforming the image  Identifying obstacles.

9. STC Robot Transformation - Example

10. STC Robot The Ant-like STC Algorithm

11. STC Robot Marking Sub-cells  Each time the robot decides to drive forward, it turns on the lights for a few seconds before proceeding, leaving a trace in one edge of the visited sub-cell.  Lights are on the back of the robot.  Traces are left only on the edge of the sub-cell.  Why not keep lights on?

12. STC Robot Finding the Parent cell  The robot needs only to  Scan the sub-cells in counterclockwise order, and  Identify the transition between unmarked and marked sub-cells.  The parent cell is in that direction. (From the STC paper.)  Upon entering a cell, the robot takes a photo of it, and finds its parent cell.  However,  The robot marks only one edge of each visited sub-cell.  This edge is shared with the neighboring sub-cell.

13. STC Robot Identifying visited sub-cells (1) Because the robot turns on the light only upon driving forward, its path in a new cell can be one of the following four:

14. STC Robot Identifying visited sub-cells (2) By examining all the possibilities where the robot is visiting a cell, we reached to the conclusion that only the 5 boxes below can be lit upon entering a cell:

15. STC Robot Identifying visited sub-cells (3)  From the above, we can conclude that:  sub-cell 1 is visited iff one of the boxes 1 and 5 is lit.  sub-cell 2 is visited iff one of the boxes 2 and 3 is lit.  sub-cell 3 is visited iff box 4 is lit.

16. STC Robot Finding the green dots  Crop the relevant section of the image.  Look for the square of pixels (of a dot’s size) that has the highest average value.  Looking for a square, instead of a single pixel allows ignoring noise.  If the maximum found value is under a threshold, then the sub-cell was not visited.  Black out the square, and look for a new dot.

17. STC Robot Rotation time calibration

18. STC Robot Angle Detection  We find 3 green dots in a row.  We calculate the slope of the connecting line.  We translate this slope into an angle.

19. STC Robot Precision  In the previous project, we only needed to identify whether or not a cell is an obstacle.  In this project, if a cell is not an obstacle cell, we need to identify visited and free sub-cells.  A sub-cell’s area is only fourth of a cell’s area.  Moreover, we have markers only on the edges of the sub-cells.

20. STC Robot Limitations  The camera we used, was not intended to work with low-light conditions, and it couldn’t identify the green light dots after less than a minute.  The system can be improved if we use different technique for identifying the green lights.  Using better camera.  Using better sensors.

21. STC Robot Results  Using a simple and cheap robot with limited sensing and computational capabilities, we succeeded in solving the covering problem using the Ant-like approach, where the robot leaves pheromones-like markers on the surface.

22. STC Robot Future Work  Using a better camera (preferably camera with a fully controlled integration time) to try covering a bigger environment.  Try to implement the whole algorithm on the robot itself, without the help of an external Computer.  Try solving more complex problems with ant-like approach solutions.

23. STC Robot Video: http://www.youtube.com/watch?v=ZkVvyl5Qmww Project paper: https://docs.google.com/document/d/1WJP4UTn463Ihyu9oQo WivwV8-8DJsqS6gEC_7ST3_Ws/edit MOVIE

24. STC Robot 2 Majd Srour, Anis Abboud Under the supervision of: Yotam Elor and Prof. Alfred Bruckstein

25. STC Robot BACKUP

26. STC Robot Finding the Parent Cell “As long as the four sub-cells of the current cell are not all marked, the robot need only to scan the sub- cells in counterclockwise order and identify the transition between unmarked and marked sub-cells. The two sub-cells bordering this transition have an exterior edge in common. This edge is necessarily the boundary edge between the current cell and its parent cell in the spanning tree.” p.84

27. STC Robot Debugging Small inaccuracies stemming from the camera, the environment, a low light from the outside, or a deviation of the robot, can cause the robot to make wrong decisions.  We saved a copy of each taken snapshot, with a timestamp on it. Thus, being able to later track the robot’s path and spot the issues.  In addition, we added a two-step initial calibration.

28. STC Robot Initial Calibration (1) Transformation Calibration:  The robot is placed on the lab floor with light on.  A snapshot is shown to the user.  The user clicks on the camera lens’ center.  The image is transformed, with the camera center’s coordinates as parameters.  The user is shown a transformed image and is asked if it’s good.  The parameters are serialized and can be reused.

29. STC Robot Initial Calibration (2) Neighbors Calibration:  The user is shown a transformed image.  The user clicks on the 12 vertices of the neighbor cells:  These are used to divide the snapshots to cells as accurately as possible.  The parameters are serialized and can be reused.