ProjectNeura
Add Historical Traces to the Map
To better analyze and improve driving strategies, we want the historical traces to be displayed as a floating layer on the map. A detailed introduction will be later added in a periodic report.
See https://github.com/ProjectNeura/LEADS/blob/master/docs/reports/2024-03-07.md.
Here's the .gpx data for the 2023 Waterloo EV Challenge downloaded from Strava. You will need to scrub the data to get the course route with and without the pit. There's some extraneous data as well but I am sure you will be able to pick that out. Here's the Strava link if you want to view the data on their app but it's easy enough to upload to google earth/maps.
See https://github.com/ProjectNeura/LEADS/blob/master/docs/reports/2024-03-07.md.
@HaydenHour Please see one of us if you have any questions about this task.
@HaydenHour Any update on this? Can you give us some more information on where you're stuck?
@ATATC @qmascarenhas Sorry I did not post sooner, I did not have access to my laptop over the weekend. It appears that the code is hitting an issue when trying to read the GPX file (specifically regarding the parse function on line 10). This is my code below:
import gpxpy
def gpx_to_custom_csv(gpx_file_path, output_csv_file_path):
# Open GPX file
with open(gpx_file_path, 'r') as gpx_file:
#Issue appears to be here with the parse
gpx = gpxpy.parse(gpx_file)
# write data in CSV format
with open(output_csv_file_path, 'w') as csv_file:
for track in gpx.tracks:
for segment in track.segments:
for point in segment.points:
# Write latitude and longitude separated by semicolon
csv_file.write(f"{point.latitude};{point.longitude},")
# file paths:
gpx_file_path = "C:/Users/hayde/OneDrive/VeC/Waterloo_Electric_Vehicle_Challenge_2023.gpx"
output_csv_file_path = "C:/Users/hayde/OneDrive/VeC/VEC2023_CSV.csv"
gpx_to_custom_csv(gpx_file_path, output_csv_file_path)
Oh, that's weird. So did it create a new file with the right formatting?
I haven't tested the file part. I replaced it with print(). But I believe write overwrites the file. There should be an appending function.
I'm gonna finish this feature (including auto corner speed marking) for the website this week.
@qmascarenhas Now you can directly import the previously recorded data onto the website and move your mouse close to the corners to check the data at that point.
@HaydenHour Sorry for the wrong instructions previously. I had some misunderstanding about CSV files. I'll provide a new handout later.
Result
@HaydenHour Take a look if you want.
from gpxpy import parse
from leads.data_persistence import CSVCollection
def gpx_to_custom_csv(gpx_file_path: str, output_csv_file_path: str) -> None:
with open(gpx_file_path, 'r') as gpx_file:
gpx = parse(gpx_file)
csv = CSVCollection(output_csv_file_path, ("t", "voltage", "speed", "front_wheel_speed", "rear_wheel_speed",
"forward_acceleration", "lateral_acceleration", "mileage", "gps_valid",
"gps_ground_speed", "latitude", "longitude"))
for track in gpx.tracks:
for segment in track.segments:
for point in segment.points:
csv.write_frame(int(point.time.timestamp() * 1000), *((None,) * 9), point.latitude, point.longitude)
gpx_to_custom_csv("Waterloo_Electric_Vehicle_Challenge_2023.gpx", "data/2023.csv")
Result
An enhanced version with inferred speeds.
from gpxpy import parse
from leads.data_persistence import CSVCollection, DEFAULT_HEADER
from leads.data import dlat2meters, dlon2meters
def gpx_to_custom_csv(gpx_file_path: str, output_csv_file_path: str) -> None:
with open(gpx_file_path, 'r') as gpx_file:
gpx = parse(gpx_file)
csv = CSVCollection(output_csv_file_path, DEFAULT_HEADER)
# prev time
t_0 = None
# prev latitude
lat_0 = None
# prev longitude
lon_0 = None
for track in gpx.tracks:
for segment in track.segments:
for point in segment.points:
t = point.time.timestamp()
lat = point.latitude
lon = point.longitude
if t_0 is None or lat_0 is None or lon_0 is None:
t_0, lat_0, lon_0 = t, lat, lon
csv.write_frame(int(t * 1000), *((None,) * 7), True, None, lat, lon)
continue
# dt
dt = t - t_0
dlat = lat - lat_0
dlon = lon - lon_0
# dy
dy = dlat2meters(dlat)
# dx
dx = dlon2meters(dlon, .5 * (lat_0 + lat))
# c
hyp = (dy ** 2 + dx ** 2) ** 0.5
# v
v = hyp / dt
# m/s to km/h
v *= 3.6
t_0 = t
lat_0 = lat
lon_0 = lon
csv.write_frame(int(t * 1000), None, v, v, v, *((None,) * 3), True, v, lat, lon)
gpx_to_custom_csv("Waterloo_Electric_Vehicle_Challenge_2023.gpx", "data/2023.csv")