ProjectNeura
VeC99 LEADS Setup for 2025 Waterloo EV Challenge
This task ownership is assigned to @fz20181223luogu - LEADS Setup and Configuration _with support from @ATATC _ @DanielS928 - Electrical Wiring _with support from @qmascarenhas _
The Task is to get VeC99 LEADS Ready for the 2025 Waterloo EV Challenge. Race day is Saturday May 24.
As @qmascarenhas mentioned to me we are going to use a Raspberry Pi or an Orange Pi 5 Plus for the VeC99. So, @ATATC can you bring one of them next Tuesday so we can start installing that on to the car?
The Raspberry Pi is literally on VeC99 and you have the Orange Pi. I'm not in charge of component purchases. All assets shall be purchased through @qmascarenhas and kept in rm217.
I was hoping you guys would figure out this plan together with minimal direction from me. But perhaps I need to provide better initial direction.
There are 3 pieces to getting VeC99 race ready:
- LEADS Configuration a) Decision: RaspberryPi or OrangePi Pro (@ATATC I thought you wanted to "sell" your Orange Pi Pro to VC - please confirm). - I propose we first test LEADS Configuration with the RaspberryPi understanding that we will have no video. - Once the above configuration is confirmed, we can start looking at the OrangePi (Pro) configuration with video. b) Confirm that we have fixes for issues identified at the Fall race: - errors relating to RaspberryPi and/or USB hubs - BATT voltage reading accuracy
- Wheel Speed Sensor Positioning (this is something I will take on with @DanielS928 and/or my SPH4Ux classes)
- Re-position wheel speed sensor for optimal Hall-Effect Sensing
- Remote Analyst Configuration
Race day is Saturday May 24. VeC99 with LEADS should be configured and verified by start of day Friday May 23. This means 3 working days (Tuesday, Wednesday and Thursday).
IMPORTANT: The racing team will be dryland practicing with VeC99 on blocks. Not sure what their schedule is yet, but that should be posted on the Edsby VeC Racing group.
@fz20181223luogu I am assuming you are now familiar with what you need to do to take care of pieces 1 and 2 above. If you have any specific questions, these should be directed to @ATATC under this issue. It is my expectation that at minimum, we will have LEADS operational on VeC99 with the currently installed RaspberryPi.
ps: @Ricky-March (Alumnus who led the design and fabrication of VeC99) is planning a visit on Tuesday after school. It might be a good time to demo LEADS on VeC99. If the racing team is using VeC99 after school, then we would have access after 4pm.
@ATATC can you please give Jerry a call - it's urgent - he's having a problem with the remote analyst connection. Everything is up and he has internet connection. Call him please so he can give you the details.
@fz20181223luogu can you please post the log files (all) for today? There should be 2 for the 2 times we powered up LEADS. @ATATC Remote Analyst stopped updating about half-way through the race.
I was in an electronic isolation room and didn't see the calls. I will look into the incident once the log files posted. We can try to reproduce the problem next week and find out the reason.
There were actually 3 log files recorded on my computer, and they are provided as follows:
2025-05-24_09-39-47.csv 2025-05-24_09-53-46.csv 2025-05-24_10-29-10.csv
I was in an electronic isolation room and didn't see the calls. I will look into the incident once the log files posted. We can try to reproduce the problem next week and find out the reason.
It's going to continue to be challenging to take VeC99 out on the road at VC so I'll put it back in the corner by the elevator adjacent to rm217 and on blocks so we can run and test as much as we need. Couple of things also need to be addressed:
- Accuracy of wheel speed calculation: looks like a period sensing accuracy/algorithm issue.
- Accuracy of the voltage sensor reading: looks like a calibration issue.
LEADS batt pack @ 20% after race. Need to check log files for total plug-in time.
I haven't changed the core of the Remote Analyst features (https://github.com/ProjectNeura/LEADS/compare/0.9.5...0.9.6). It might simply be that @fz20181223luogu 's laptop isn't powerful enough. Running for 100 minutes can be challenging for laptops with a small memory capacity as the dashboard stores massive amounts of data in the browser.
@fz20181223luogu @qmascarenhas
dataset: "data/main.csv"
inferences:
repeat: 1 # default: 1
enhanced: false # default: false
assume_initial_zeros: false # default: false
clear:
- speed
methods:
- speed-by-gps-ground-speed
jobs:
- name: Task 1
uses: bake
[Level.INFO: 1] [run] [2025-05-28 13:49:45.184614] Baking Results
Baked 95287 / 95287 ROWS
Baking Rate: 100.00%
Skipped Rows: []
Start Time: 2025-05-24 10:29:12
End Time: 2025-05-24 12:08:20
Duration: 99 MIN 8 SEC
Distance: 62.29 KM
vₘᵢₙ: 0.00 KM / H
vₘₐₓ: 28.00 KM / H
vₐᵥᵍ: 37.70 KM / H
GPS Hit Rate: 100.00%
GPS Skipped Rows: []
Min Video Latency: N/A
Max Video Latency: N/A