WORK ACCOMPLISHED:
Landed on a solidified architecture for controller to board communication.
The flow will be as follows.
In the event of normal riding, our flow will be input from user on webapp -> bluetooth communication to raspberry pi -> raspberry pi to VESC communication of signals -> VESC to motors
In the event of the “return to me” feature, our flow will be input from user on webapp -> bluetooth communication to rasberry pi -> input from LiDAR sensor to raspberry pi -> path planning calculation on the web app -> directions sent to raspberry pi -> directions sent to VESC -> directions sent to motors.
Part Selection:
- Landed on an almost-final version of the parts we will be employing.
- VESC: https://flipsky.net/products/dual-fsesc4-12-100a?variant=9022834638908¤cy=USD&utm_medium=product_sync&utm_source=google&utm_content=sag_organic&utm_campaign=sag_organic&gad_source=1&gclid=CjwKCAjw0aS3BhA3EiwAKaD2Zd5iglwJYeZaNOX5BAR4TkeQuqevF_zDt3az7iHfjzF1aVxhLHioOhoC1rkQAvD_BwE
- 2x 5055 motors: https://www.amazon.com/Hobbyhh-Brushless-5055-400KV-Efficiency-Aircraft/dp/B09BYVKZ5W/ref=sr_1_2?crid=F1IOMBUFCJGT&dib=eyJ2IjoiMSJ9.YIcSePfgaVsN6txie0egmwsARIbLiLZgkDep_XMtgkBZ48uyUE-Bgfs2PnUpL4Tzld0_cKCw4WfuliYydziJmzvCVVH7j8HK2DLiDKXKFYpvkwgklkCSKVAfWKhmS51yGr04LtxTChMIHHGz2P-7HteXbZ5jz4_VQE1rC8LA2kruijlqTRfQe7ytiWh-B36U59Fcbog-XeS5GYgmFXH0S-JPCSWt8fuD1-8DCrpFwQ4eO9xw7Iw0vmxQUxG7mOkWmbkssPNsyoN1JPxpTmU12v4wypoSK778X7mkfMDxETg.V86EyqCGVVZR9GmM6c0lQvJXmOMAs4k0VaUqd1ns8YU&dib_tag=se&keywords=5055+motor&qid=1726985746&s=toys-and-games&sprefix=5055+motor%2Ctoys-and-games%2C289&sr=1-2
- The build above minimized spending but allows us to meet the requirements we specified in our abstract and in our proposal presentation
Discussions:
- More info about architecture: The VESC turns out to be necessary due to power consumption concerns: the raspberry pi is unable to regulate and control the amount of voltage and current needed to power the motors. As a result, we cannot circumvent the VESC and must use a rbpi alongside a VESC.
- On the topic of belt driven vs hub motors, I have elected to go with belt motors as we likely cannot sustain the cost of hub motors. Above is the candidate I have chosen. We are sacrificing ease of configuration for cost and potentially slightly more torque.
- There are a variety of open source path planning algos but for our sake, I think ML may be unnecessary as we plan to essentially implement a simple “back up and reroute” algo. As a result, I have elected to largely ignore available path planning algos.
PROGRESS:
I am on track as I was supposed to explore path planning algos and assist with part planning and both of those items have been accomplished.
NEXT WEEK’S DELIVERABLES:
Next week, I aim to complete the following:
- Have a more concrete and mathematical approach to the path planning algo including GPS movement from point A to point B, object sizing algorithm ideas, and ideating on how much to back up in the case of an obstacle occurrence.
- I want to get my team approval on my parts selection and submit it to the TA for ordering
- I will also help prepare design presentation slides and write speaker notes to help whoever presents
- I will also continue to coordinate with professors and our TA as well as make relevant updates to the slack and website