Integrated System Build & Test
Table of Contents
Team Vision for Integrated System Build & Test PhaseSummarize:
- What did your team plan to do during this phase?
During this phase our team wanted to finish up the complete system build and accomplish the tests associated with our competition robot. Overall this included ordering and adding timing belts instead of using the V belt drive system currently set in place. After this was to occur we were planning on getting the code squared away and ensure that it worked properly and then complete the associated tests. Additionally, we wanted to get some rough copies completed for our paper, poster and the beetle weight bot which we plan on presenting at ImagineRIT on April 28th, 2018.
- What did your team actually accomplish during this phase?
Since our team has been using self funds and then receiving reimbursements with a long lead time we decided to utilize the Simone Center's purchase system, which since it was the first time we used it took a while. Ultimately we ended up receiving the pulleys although a little later than we wanted. In addition to this we also ordered the new timing belts, which ultimately were too large for the ordered pulleys. To remedy this we ended up 3D printing new pulleys to be able to utilize the timing belt ordered. Since we were not able to get the robot functioning as desired mechanically we were unable to complete any of the final level tests and software checks. Looking at the progress made on the paper, poster and models of the beetle weight bot, we ultimately created rough copies of the paper, poster and models. We will need to refine these for ImagineRIT and the end of the class. These are mentioned more below.
Test Results Summary
Updated Testing document. Minimal Progress due to shipping and finance delays on the part of the Simone Center.
Subsystem Integration represents a significant risk, as the robot has been disassembled and must be reassembled. However Timer and communication concerns have been resolved. Additionally, The weight of the bot is no longer an issue as Motorama has passed.
Imagine RIT/Final Deliverable Updates"Featherweight Bot"
The Magnetometer Sample rate and communication speed were tested by toggling a pin on an off based on the status of the magnetometer communication. Upon the falling edge of the magnetometer's data ready pin, the teensy triggered an interrupt which set a flag and toggled a pin on. This Flag allowed the main loop to sample from the magnetometer and then toggle the pin back off.
The magnetometer was setup to sample at 200Hz. The microcontroller handled the communication in 1.788ms. This may be reduced slightly by ignoring two of the accelerometer axes and one of the magnetometer axes. However it may be advantageous to use that data to account for the tilt of the PCB within the robot.
Beetle Weight Bot
Motors were Selected for the Beetleweight version of the robot by using the askaaron reference document.
ESC's were selected based on their compatibility with SimonK firmware and their voltage and ampacity ratings.
Batteries for the 3lb bot were chosen based on voltage, ampacity, capacity, size and cost.
Looking at the beetle weight bot we have created a similar design to the full scale bot, although made it less lethal by:
- Removing the teeth
- Using a 3D printed case
- Decreasing the weight/making it smaller
- Decreasing the power
The reason for this is since we do not want anyone to be injured during or after ImagineRIT.
Other changes which we have implemented include:
- Using direct drive since it is a smaller and less powerful bot
- Making the compartment covers clear so that the electronics could be seen (aesthetics for the audience and possible kits)
Below are a couple of drawings showing the assembly of this bot:
For both pictures, the parts are as follows:
- Purple = 3D printed chassis
- Red = 2 ESC stacked one above the other
- Black = 2 Lite Flite Wheels
- Dark Blue = LiPo Battery
- Clear = Protection Covers
- Black Center = Electronic Cluster
Poster The poster can be seen below. Click for a larger image.
Problem TrackingBelow is our problem tracking.
Technical PaperBelow is a link to our technical paper on our google drive.
Functional Demo MaterialsInclude links to:
- Presentation and/or handouts
- Notes from review
- Action Items
Plans for next phase
Use of the Magnometer for Heading EstimationEarly software tests show that reading the magnetometer on the LSM303DLHC chip we have can be performed at a relatively fast pace. As a result, it may be possible to create a heading estimation with only the magnetometer readings, if only at lower speeds. Doing so comes with a few caveats, however.
- Firstly, while the magnetometer readings may be frequent enough for uses regardless of our original accelerometer method, it's still not frequent enough to be real time. This means that some form of prediction will need to be made from the heading readings. This could be simple interpolation over the previous three or more readings, or it may need to be more complex. Research and experimentation will need to be done before a conclusion can be made.
- Secondly, a method of tilt compensation will need to be made so that the wobble of the robot can be accounted for. A method of conversion for magnetometer and accelerometer readings into pitch, roll, and yaw has documentation exists from the distributor of the sensor, and other adequate documentation exists for similar methods. This makes it an issue of implementation and testing. However, further difficulties may result from the g-force experienced by the accellerometer on the LSM303, meaning they may be unusable for tilt compensation.
- Thirdly, if the magnetometer proves inadequate alone for determining the heading, then we will need to combine the readings with that of the accellerometer as was our original plan. Given the slow speed of reading and calculating the two sensors, we will need a way of speeding up the process to keep a high speed of rotation for our robot.
- Develop an interface library for reading the LSM303DLHC and calculating heading, including tilt compensation
- Continue Software Development
- Improve translational logic
- Assist in testing the original bot and the mini-bot
- Finalize Poster
- Finish modeling the beetle weight robot(wiring channels, details, etc.)
- Create 3D printed/machined parts
- As parts come in assemble
- More robot testing
- Electrical Work for Beetleweight bot when parts arrive
- Mechanical Assembly Help
- Software testing help
- Help Re-Assemble 30Lb bot for testing
- Help With Software Issues
- Continue working on technical paper rough draft
- Creat video for our Imagine RIT booth
- Mechanical Assembly Help
- Help Re-Assemble 30 lb bot for testing
- Test robot both 30 lb and beatleweight bots
Home | Planning & Execution | Imagine RIT