Add -v,--version flag for checking installed version

I would like to break out the path simulation code into a library API called by the CLI. This will allow for tests around core functions, as well as the ability to use the GPS physical simulation functionality in custom use cases, such as creating paths from a pre-existing OD pair dataset.

I tried to run the docker file but I got "executor failed running [/bin/sh -c /trip-simulator/node_modules/osrm/lib/binding/osrm-extract" error using node:11 in dockerfile. Would you please let me know how I can resolve this issue?

Trying to use the tool to simulate trajectories with a dataset of real OD data.
Is it possible with the current command line arguments?
If yes how should I prepare my OD data?

Add a flag to set the start datetime.

MDS >= 0.3 uses epoch milliseconds instead of epoch second floats. Let's make the change here.

I received a bug report that occasionally the first timestamp is not populated or possibly null.

I would like to expose all variables that might be impacted by the type of driver and hardware being simulated, with the goal to support multiple types of vehicles. The following represent medians for normal distributions used by the simulation:

• horizontal accuracy
• accuracy variance
• probe frequency
• frequency variance
• routing type (car, bike, ped, other)
• wait between trips
• distance between trips (eg: 0 for dockless bikes and scooters / 0.5km for taxis)
• trip distance
• speed deviation (+- % duration from routed ETA prediction)
• fleet size
• trips per vehicle per day

Move away from the hardcoded makefile to a CLI. This will make it easier to generate data for multiple areas.

trip-simulator --config ./taxi.json --osm ./nyc.osm.pbf --out ./data
# creates ./data dir
# extracts & contracts graph from nyc.osm.pbf
# outputs trips to ./data/trips.json
# outputs status_changes to ./data/status_changes.json

I'm having trouble installing trip-simulator using the npm install -g trip-simulator command.

System details:-
Distributor ID: Ubuntu
Description: Ubuntu 18.04.1 LTS
Release: 18.04
Codename: bionic

node -v
v12.6.0

npm -v
6.9.0

npm install -g trip-simulator
/usr/bin/trip-simulator -> /usr/lib/node_modules/trip-simulator/src/cli.js

[email protected] install /usr/lib/node_modules/trip-simulator/node_modules/osrm
node-pre-gyp install --fallback-to-build=false || ./scripts/node_install.sh

node-pre-gyp WARN Using needle for node-pre-gyp https download
node-pre-gyp ERR! install error
node-pre-gyp ERR! stack Error: EACCES: permission denied, mkdir '/usr/lib/node_modules/trip-simulator/node_modules/osrm/lib/binding'
node-pre-gyp ERR! System Linux 4.9.125-linuxkit
node-pre-gyp ERR! command "/usr/bin/node" "/usr/lib/node_modules/trip-simulator/node_modules/osrm/node_modules/.bin/node-pre-gyp" "install" "--fallback-to-build=false"
node-pre-gyp ERR! cwd /usr/lib/node_modules/trip-simulator/node_modules/osrm
node-pre-gyp ERR! node -v v12.6.0
node-pre-gyp ERR! node-pre-gyp -v v0.12.0
node-pre-gyp ERR! not ok
EACCES: permission denied, mkdir '/usr/lib/node_modules/trip-simulator/node_modules/osrm/lib/binding'
mkdir: cannot create directory 'build': Permission denied
npm ERR! code ELIFECYCLE
npm ERR! errno 1
npm ERR! [email protected] install: node-pre-gyp install --fallback-to-build=false || ./scripts/node_install.sh
npm ERR! Exit status 1
npm ERR!
npm ERR! Failed at the [email protected] install script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR! /root/.npm/_logs/2019-07-09T16_44_36_224Z-debug.log

I have changed the npm global installation location to a local directory. The installation still fails. Appreciate all the help.

We should pre-build some samples and host on s3 for easier distribution.

I would like to try simulating the OD pairs in addition to the paths. This would allow for the removal of the dependency on real world TLC data, and allow simulations to be created in other regions or even globally. Pairs could be generated probabilistically based on node density, which can be calculated from the OSM data the tool already needs for operation.

https://github.com/CityOfLosAngeles/mobility-data-specification/tree/dev/provider#status-changes

Simulate status_changes from a simulated trip feed.

We're using the chance lib under the hood for providing pseudo-random values and distributions throughout the model. This lib allows a fixed seed string to be provided for reproducible output. This is useful for reproducing reported errors and behavior from the community, and will make it easier to write unit tests with meaningful assertions.

We currently support default configs for cars, bikes, and scooters, however, I often find myself needing to tweak various values to test edge cases. For example, I may need to simulate a scooter company that has more frequent breakdowns than usual or a taxi company that uses lofi GPS units.

This proposal would add support for a file path as the config, with the existing keywords maintained.

trip-simulator --config bike
trip-simulator --config ./bike-with-faulty-wheels.json

Unavailable vehicles should eventually be removed from the street. This should be a configurable parameter to define the distribution, and each vehicle should spawn with a different time to pickup, weighted to that distribution.

Agent GPS can use surrounding building density and height to weight horizontal accuracy towards higher skew. The data is already available to the algorithm, since it is used in OD selection, but may require an additional quadtree aggregator at a higher zoom with average height aggregated as well.

Timestamps should be interpolated from route leg ETAs to provide realistic speeds with some amount of variance per vehicle, per trip, and per leg.

In each config, an attribute frequency is set, but does not appear to be used elsewhere in the code:

Is this an unused attribute or, if it is used, how/where?

Hi I'm trying to run a simple example over Stockholm, Sweden and keep running into a heap space issue.

node ./src/cli.js --max-old-space-size=12192 --config car --pbf stockholm_sweden.osm.pbf --graph ../../data/sweden-latest.osrm --agents 2 --iterations 400

<--- Last few GCs --->

[72008:0x102803800]    53467 ms: Mark-sweep 1407.5 (1449.4) -> 1407.5 (1449.4) MB, 1140.7 / 0.0 ms  allocation failure GC in old space requested
[72008:0x102803800]    54866 ms: Mark-sweep 1407.5 (1449.4) -> 1407.5 (1433.4) MB, 1398.3 / 0.0 ms  last resort GC in old space requested
[72008:0x102803800]    56185 ms: Mark-sweep 1407.5 (1433.4) -> 1407.5 (1433.4) MB, 1319.1 / 0.0 ms  last resort GC in old space requested


<--- JS stacktrace --->

==== JS stack trace =========================================

    0: ExitFrame [pc: 0x2685a130427d]
Security context: 0x2dfbe80206a9 <JSObject>
    1: /* anonymous */ [<redacted>/node_modules/trip-simulator/src/simulation.js:~44] [pc=0x2685a13698ba](this=0x2dfb4fe1d251 <JSGlobal Object>,item=0x2dfbd1a9b2c1 <Object map = 0x2dfb28eefe41>)
    2: arguments adaptor frame: 3->1
    3: forEach(this=0x2dfbbd605569 <JSArray[2405817]>)
    4: /* anonymous */ [<redacted>...

FATAL ERROR: CALL_AND_RETRY_LAST Allocation failed - JavaScript heap out of memory
 1: node::Abort() [<redacted>/nodejs/10.0.0/bin/node]
 2: node::FatalTryCatch::~FatalTryCatch() [<redacted>/nodejs/10.0.0/bin/node]
 3: v8::internal::V8::FatalProcessOutOfMemory(char const*, bool) [<redacted>/nodejs/10.0.0/bin/node]
 4: v8::internal::Factory::NewFillerObject(int, bool, v8::internal::AllocationSpace) [<redacted>/nodejs/10.0.0/bin/node]
 5: v8::internal::Runtime_AllocateInTargetSpace(int, v8::internal::Object**, v8::internal::Isolate*) [<redacted>/nodejs/10.0.0/bin/node]
 6: 0x2685a130427d
 7: 0x2685a13698ba
Abort trap: 6

I've tried with node 8.5 and 10 (OSRM didn't work with 12). Do you have any suggestions on which node version to use?

Br,
Mats

I can see in the src how each trip in trips.json is assembled:

Where does the 1000 constant in the denominator come from? I think it's related to the stepSize from the simulation being hardcoded to 1000:

If so, are the trip_duration and trip_distance values the in seconds and meters, respectively?

I ask because I'd assume the time would converted /60 (min) /360 (hrs), so wondering where the 1000 is coming from.

Same question roughly with distance value.

Add linter for style consistency.

👋hey @morganherlocker - thanks for this amazing tool 🙇

Wanted to report a small issue that I've observed: the change event time for user_drop_off is always greater than the trip end_time. This appears to be happening due to this code section: https://github.com/sharedstreets/trip-simulator/blob/master/src/agent.js#L200-L232

if (this.simulation.time >= this.next) {
...
  if (this.changes) {
  ..
    event_time: this.simulation.time
  ..
  }
if (this.trips) {
  const trip = {
    ..
    end_time: this.next,
    ..
}

Apart from the values being different, the end time is also a floating point number instead of an int.

I've locally tested a fix

    end_time: this.simulation.time,

and it works - but I wanted to check with you incase there was a reason to log them differently?

Generate a random trip ID for each trip using the chance UUID generator.

Seeing some unexpectedly jagged traces when I simulate Dupont Circle in DC using the car profile.

The noise simulation seems like a reasonable approach to the problem, though not a particularly realistic one (looking at some actual telemetry I see a lot more parallelism than random jitter, perhaps due to either environmental RF echoes or the tendency of the location services black box trying to smooth its output by keeping heading relatively constant).

Assuming realism isn't a particular goal, I'll throw out two possible changes. Are there other ways of tackling this?

• having the location error as a cumulative vector that is adjusted at each step along a trip rather than a random error added at each step
• multiplying the error by a function of between-node distance to diminish its effect around circles

I'm also curious why the heading for the noise is normally distributed.

Happy to take a crack at a PR if any of this sounds like a good idea.

Running on Ubuntu 16.04 LTS,

Is the problem npm version or my OS?

$ cd trip-simulator/
$ npm version
{
'trip-simulator': '3.0.1',
npm: '6.14.13',
ares: '1.17.1',
brotli: '1.0.9',
cldr: '39.0',
icu: '69.1',
llhttp: '2.1.3',
modules: '83',
napi: '8',
nghttp2: '1.42.0',
node: '14.17.1',
openssl: '1.1.1k',
tz: '2021a',
unicode: '13.0',
uv: '1.41.0',
v8: '8.4.371.23-node.67',
zlib: '1.2.11'
}

$ npm install -g trip-simulator
npm WARN deprecated [email protected]: core-js@<3.3 is no longer maintained and not recommended for usage due to the number of issues. Because of the V8 engine whims, feature detection in old core-js versions could cause a slowdown up to 100x even if nothing is polyfilled. Please, upgrade your dependencies to the actual version of core-js.
npm WARN checkPermissions Missing write access to /usr/lib/node_modules
npm ERR! code EACCES
npm ERR! syscall access
npm ERR! path /usr/lib/node_modules
npm ERR! errno -13
npm ERR! Error: EACCES: permission denied, access '/usr/lib/node_modules'
npm ERR! [Error: EACCES: permission denied, access '/usr/lib/node_modules'] {
npm ERR! errno: -13,
npm ERR! code: 'EACCES',
npm ERR! syscall: 'access',
npm ERR! path: '/usr/lib/node_modules'
npm ERR! }
npm ERR!
npm ERR! The operation was rejected by your operating system.
npm ERR! It is likely you do not have the permissions to access this file as the current user
npm ERR!
npm ERR! If you believe this might be a permissions issue, please double-check the
npm ERR! permissions of the file and its containing directories, or try running
npm ERR! the command again as root/Administrator.

npm ERR! A complete log of this run can be found in:
npm ERR! /home/michael/.npm/_logs/2021-06-22T23_52_43_665Z-debug.log

So tried sudo:
$ sudo npm install -g trip-simulator
npm WARN deprecated [email protected]: core-js@<3.3 is no longer maintained and not recommended for usage due to the number of issues. Because of the V8 engine whims, feature detection in old core-js versions could cause a slowdown up to 100x even if nothing is polyfilled. Please, upgrade your dependencies to the actual version of core-js.
/usr/bin/trip-simulator -> /usr/lib/node_modules/trip-simulator/src/cli.js

[email protected] install /usr/lib/node_modules/trip-simulator/node_modules/osrm
node-pre-gyp install --fallback-to-build=false || ./scripts/node_install.sh

node-pre-gyp ERR! install error
node-pre-gyp ERR! stack Error: EACCES: permission denied, mkdir '/usr/lib/node_modules/trip-simulator/node_modules/osrm/lib/binding'
node-pre-gyp ERR! System Linux 4.15.0-142-generic
node-pre-gyp ERR! command "/usr/bin/node" "/usr/lib/node_modules/trip-simulator/node_modules/.bin/node-pre-gyp" "install" "--fallback-to-build=false"
node-pre-gyp ERR! cwd /usr/lib/node_modules/trip-simulator/node_modules/osrm
node-pre-gyp ERR! node -v v14.17.1
node-pre-gyp ERR! node-pre-gyp -v v1.0.5
node-pre-gyp ERR! not ok
EACCES: permission denied, mkdir '/usr/lib/node_modules/trip-simulator/node_modules/osrm/lib/binding'
Needs cmake to build from source
npm ERR! code ELIFECYCLE
npm ERR! errno 1
npm ERR! [email protected] install: node-pre-gyp install --fallback-to-build=false || ./scripts/node_install.sh
npm ERR! Exit status 1
npm ERR!
npm ERR! Failed at the [email protected] install script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

npm ERR! A complete log of this run can be found in:
npm ERR! /home/michael/.npm/_logs/2021-06-22T23_55_37_670Z-debug.log

First line of the TLC data contains a CSV header, which causes some error noise when the date parsing is attempted.

Let's add timestamp interpolation to traces, like we do for MDS style trips. This will bring us into compliance with the TraceJSON specification.

Add screenshots to readme demonstrating noise simulation.