Improve precision of the lane position detector

[turbokadi]

Related to #31

[theomarzaki]

Hello @J4BB3R ,

I hope you are well !

I noticed that some of the code is results in an error.

I wondering if this may be a result of only having one car on the road (and database). I don't know if that impacts your code heavily.

[turbokadi]

@theomarzaki Hum, Ok I've found the bug, two bugs actually. The big number in the the stack is due to the non loss priority of the compiler. I've use {} to prevent narrowing, so when the compilation runs, gcc thought that the below calculus cannot be negative because node.size and max are declared as unsigned long. So it's always 'gps.nodeId - spread' who is executed and it may produce a negative index -> max 18446744073709551615 - 6 = the big number in the stack trace. that's the first one. The second one, is basically because I haven't reread my ternary statement properly. The fact I'm doing max - and not max + changing the sign and do actually what I don't want to do. After all it seems that my implementation need a serious improvement in precision for detecting the position on the road, it don't affect the faker that much but it can screw the result when we are far from the merging lane, because the points between the lanes begins to de-synchronize (not aligned). So I'm working on it. I've changed and pushed the fix. Have a nice day.

[theomarzaki]

Ahh, Perfect,

Its great your finding these bugs !

But unfortunately, the fixes did not fix the underlying bug.

I will comment out the implementation in create Trajectories to prevent crashes in master,

Would it be possible to push the upcoming changes to its branch ?

Thank you very much ,

I hope you have a nice day as well !

[turbokadi]

That's odd, it's the same error ?

[theomarzaki]

@J4BB3R yes unfortunately,

to check , you can ./runner with the code and send 1 line

from http://90.84.192.152:8080/data-fusion to the v2x gateway address (In TO_config)

[turbokadi]

Hi @theomarzaki, My bad ! I thought the error came from your own instance. The code wasn't available on the platform when you've tested, it have been updated on the platform only 4 hours ago. I haven't forced the pipeline myself.

You have modified the master before the trigger ... So it's your code now. I'm quite sure my code works. I may force the pipeline on the master if you remove your comments (I can do it, if you don't have the time).

[theomarzaki]

Hello @J4BB3R ,

I think I may a bit confused. Could you push the changes that you think works on a new branch, and I will go through that branch before merging into master. Also , the trigger is not a problem here, since I was testing on my local instance your code , I didn't take it from the pipeline .

[turbokadi]

@theomarzaki Just to check, that's mean you're compiling the code on your computer, you're not using the docker images on the server. If it's the case can you give me your traces please, to look what's wrong ? I can't reproduce it anymore.

[theomarzaki]

Yes @J4BB3R , that is correct !

below is the trace obtained. Note that the error is actually very similar to the error received above.

Unfortunately, the application stack is not working so there are no messages being sent.

When the stack is up and running I will capture the trace and send it over !

[tigroo31]

Hi the workers! The TO is a very critical composent for the 5GCar project, you know ;). Please : 1- Do not commit directly in the master but use a pull request explaining how to test your new feature / the fixbug. 2- Please test at least one time locally with the last injector using the Montlhéry 3 test suite , to be sure that the complete demonstration continue to work. On this last rush for June, we must increase the reliability of the component, testing more in branch before to deploy to everybody. Let’s talk about this on next meeting tomorrow ;)

Le 14 mai 2019 à 13:51, Omar Zaki notifications@github.com<mailto:notifications@github.com> a écrit :

Yes @J4BB3Rhttps://github.com/J4BB3R , that is correct !

below is the trace obtained. Note that the error is actually very similar to the error received above.

Unfortunately, the application stack is not working so there are no messages being sent.

When the stack is up and running I will capture the trace and send it over !

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/theomarzaki/TrafficOrchestrator/issues/46?email_source=notifications&email_token=ALMTQ3RS4KHYSQFHRTWPVELPVKRVZA5CNFSM4HMNSRR2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVLHJ4A#issuecomment-492205296, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ALMTQ3WLGJA7EMIGVFNYBJTPVKRVZANCNFSM4HMNSRRQ.

---

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration, Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law; they should not be distributed, used or copied without authorisation. If you have received this email in error, please notify the sender and delete this message and its attachments. As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified. Thank you.