Final Deliverable (Week 10)

The project has been completed.  Maze completion was a success, as the group decided to add a a baseboard to the bottom of the maze for added stability.  The final maze can be seen below.

Final iteration of the maze

The robot design was kept the same as in previous weeks, with no alterations.  Unfortunately, the robot was not able to successfully navigate through the entire maze and extract an object, but it was able to navigate the maze.  It was also capable of capturing an object separately, which demonstrates that the coding was successful in getting both Mindblocks to work together.  Videos of the robot's capabilities can be seen below:

Robot Maze Navigation - Final

Object Capture Demonstration

The final report and presentation PowerPoint have been uploaded to the home page of this site.

Week 8 Into Week 9

This week, the maze walls were completely constructed, as shown in the images below:

Constructed Maze:  Full View

Constructed Maze:  Example Joint View

Using duct tape, each piece of cardboard wall was adhered together in four separate points: top inside, bottom inside, top outside, and bottom outside.  This was done in order to ensure a stable and sturdy construction that would not fall apart from various wear such as robot collisions during testing or, ultimately, the outside winds when transporting the maze outside.  Currently, due to the instability and excessive flexibility of the outside walls, there are plans to create a cardboard bottom floor for the maze walls to attach to, thus allowing for more stability and proper spacing of the walls.

UPDATE:  Due to the immense size of the maze and difficulty of transportation, the robot will not need to be transported for its final demonstration.  Instead, a video will be shot of the robot navigating the maze, which will be posted in an update on this website.

Robot testing has been on its way.  Currently, the robot is able to navigate through the walls of the maze.  Object detection is also on its way, with the robot's magnetic sensor able to recognize the magnetic object that will be extracted from the maze.  

Objectives Over the Next Week:

Further test and develop robot to have a completed deliverable

Week 7 Into Week 8

This week the maze went under heavy construction.  Duct tape was acquired and roughly half of the maze was constructed in the lab using the cardboard that was cut out.  All testing of the robot's NXC coding has been unsuccessful.  No further progress has been made.

Objectives Over the Next Week:
Complete maze by end of next class
Robot needs to be functional to complete maze
Begin presentation preparation

Week 6 into Week 7

Maze construction has finally begun this week.  As seen in the figure below, the cardboard was delivered to the group cut into 22 sheets of 8 x 60" strips.

Initial State of Cardboard Strips

The strips were then cut into pieces as follows for the walls of the maze:

12 sheets of 8 x 11.25"
9 sheets of 8 x 22.5"
5 sheets of 8 x 33.75"
1 sheet of 8 x 45"

For the outside walls of the maze, 4 strips of 8" x 2'6" were cut and taped onto 4 existing 8 x 60" sheets to create strips of 8" x 7'6".  All of the strips can be seen below.

 Cardboard Strips Cut into Proper Sizes:  Top View

 Cardboard Strips Cut into Proper Sizes:  Side View

The group has decided to construct the maze in the lab itself in order to avoid conflict with the elements  (mainly the high winds of Philly) when trying to transport the entire maze between our dorms and the lab.  The group initially decided that hot glue would be used to connect the cardboard pieces of the maze; however, upon arriving at the lab and attempting to construct with hot glue, it proved to be ineffective and impractical.  Currently, the group has decided that duct tape will used instead, as it will provide a more reliable adhesion between the sheets of cardboard.  The sheer size of the maze is exemplified by the figure below.

Outside Walls of the Maze Propped Up

The group will store the maze at the lab due to the impracticability of attempting to transport an object of such sheer size and frail material through the unpredictable high winds and other elements of Philadelphia.

The NXC code is coming along well.  An example of the code can be seen below.

Small Portion of NXC Coding for Robot Testing

Currently, the group is in the beginning stages of testing out the robot with the code.  Tests are still a bit rough, but much progress is being made.

Week 5 Into Week 6

Progress is a bit behind schedule this week, but the team has planned out the steps that will need to be taken over the next week, and it will be easy to keep on track.  It has been decided that the maze will definitely be constructed out of cardboard, and after calling around to multiple stores in the area, the team has decided to use 40"x60" sheets of cardboard from Blick Art Materials(http://www.dickblick.com) to construct the maze.  Maze construction will begin as soon as the cardboard is acquired from the store.  It has been decided that the group will not utilize the sides of the Arena in the lab as the perimeter for the maze - instead the maze perimeter will be attached to the maze itself and will be constructed out of the same cardboard material as the rest of the board will be.  This will allow for the team to have control over all walls of the maze.  The construction of the robot is complete, as shown by the images below. 

Current Robot State - Front View

 

Current Robot State - Left Side View

Current Robot State - Left Side View

Currently, there are two ultrasonic sensors on the sides, one color sensor on the left side, and two magnetic sensors in the front.  The gripper mechanism has been modified to be longer and wider.  Currently, the NXT-C robot coding has begun, however it is just in its beginning stages.  Testing of the robot with the coding will begin over the next week.

Objectives over the next week:
Purchase maze materials
Begin maze construction
Continue NXC programming
Begin code testing with robot

Week 4 Into Week 5

This week the group mainly focused on different coding methods.  Different methods discussed such as bluetooth commands, master/slave robot, and remote control. The remote control method, where the group would guide the robot through the maze remotely, is not up to par with the design project standards. Synchronization between the servo motors is being researched. Robot sensors are being tested on the Bricx CC program. The sensors  being used still include the color sensor, ultrasonic sensors, touch sensors, and magnetic sensors.  The group also decided that using plywood to construct the maze would be highly impractical and expensive, thus, the group is currently deciding between cardboard and styrofoam to use as wall material for the maze.  Other than sensor work, not much more progress has been made on the robot itself.
Objectives over the next week:
Acquire maze materials
Begin maze construction
Sensors attached to robot
Servo motors functional
Develop coding to run both Mind-blocks at the same time  

Week 3 Into Week 4

This was a productive week for the group.  Seth designed the layout and dimensions for the maze, which Lee then created in AutoCAD (see screencap in Figure 1). 

 Figure 1:  Maze Design with Simple Dimensions.

The leftmost circle will be the starting point, while the

rightmost circle will be the ending point.

 

It was decided that the maze will utilize the Arena in the lab in Bossone 624 (Figure 2) as its enclosing border, which has a length of roughly 7'6" on all sides.  For this to be effective, all of the obstacles that are currently in the Arena, such as the miniature train wreck, will be taken out and all downed fences will be placed in their upright position.  The fence perimeter will serve as the perimeter of the maze, while the maze itself will feature paths that are roughly 11.25" wide.  Currently, the plan is to use plywood to construct the walls of the maze.

Figure 2: Lab Arena

 

Kyle, our programmer, has been continuing to learn the NXC coding language this week.  He is currently reading through the tutorials concerning the sensors as well as the motors.  Meanwhile, Bruce and Michael have been designing and constructing the NXT robot.  It has been decided that the robot will utilize two NXT bricks to perform its task of driving through the maze and retrieving the object.  Currently, the plan is to use three ultrasonic sensors placed in different directions on top of the robot in order to sense the walls of the maze, while a gripper, touch sensor, and magnetic sensor will be used to retrieve an object (Figures 3, 4, and 5).  Currently, the one ultrasonic sensor, one touch sensor, and one magnetic sensor have been constructed onto the robot.

Figure 3:  Robot In Its Current State (Front View)

 

Figure 4:  Robot In Its Current State (Side View)

 

Figure 5:  Robot In Its Current State (Rear View)

 



The group is still planning and designing which sensors will be connected to which brick.  The means of communication between the two bricks (if any) also needs to be decided upon.  The implementation of a color sensor is also in talk amongst the group, as the plan is to have the robot follow certain markers along the maze in order to discern where its position is inside of the maze.


Objectives over the next week:
Continue robot construction to have functionality
Finalize maze design plans and begin maze construction planning
Begin with NXC coding
Decide on sensor implementations onto bricks