049 Novel Traffic Control Method

[@wktodd]

Re: your "safe" right turns. Are you sure the zoomie behind the turning one will be able to see it at an angle? If not, perhaps add a cornerhcube reflector to the rear of each zoomie .

[@MakingStuffwithChrisDeHut]

Good morning wktodd, I am not sure of anything on this experiment - LOL!!!!! The last month I have been battling some strange issues regarding positional feedback that did not exist before. For this particular issue, my concern for "safe" has to do with the Zoomie making the turn. The Zoomie behind is somewhat safe. As for a reflector, I don't think that would work with the sensor as it is actually measuring the "size" of the light beam to determine distance. I have not tried it but I do know this, if the surface is not large enough, flat enough, and ideally white enough, it does not work very well. I found this out with the "parking barriers". The fist designs I created were about 20% smaller than the ones I have now. I made them bigger because I could not get reliable distance readings. Cheers! Chris

[@wktodd]

Hi Chris, your blockers have the same basic function as traffic lights. Why not just add traffic signal logic to each junction ( many of them here have stationary vehicle detectors so the lights stay green most of the time)

[@MakingStuffwithChrisDeHut]

Hello again! That is where I started with the original concept and may end up back at it. Currently I am just letting my imagination wander around to see what other methods may be possible. The more I have been looking into this ( handling collision avoidance), the more I am seeing that it is a very complex issue to deal with based on the limited sensing systems the Zoomie has available. As of today (10/19/2024) my thinking is that I am going to just keep thinking on the issue for now. I hope to start having the system running in automatic mode within the next month or so (everything takes SOOOOO long to do at this stage). With that, I can get a better understanding of what and where the highest potentials are for crashes and narrow down some potential solutions. Thanks VERY much for the feedback and suggestions! Cheers! Chris

[@peterwest1158]

Hi Chris, I am following the series with great interest. The sprites seem to be a useful mechanism, I look forward to later episodes. One thing I have wondered about is a global parameter to tell the Zoomies whether they are left or right handed. Would make things easier for those of us who drive on the left hand land to visualise the system. Would this have impacts on the Zoomie world eg the placement of the magnets etc? Your thoughts. Peter

[@MakingStuffwithChrisDeHut]

Good morning Peter!!!! Let me answer your question first... No. As you suspected, the mechanical interaction between the Zoomie and the placement of the magnets would prevent that. Unfortunately, this past month of testing may be proving my concept of using the magnets is not reliable as well. Not actually the magnets, but their placement and my implementation of the idea. The problem I cannot fix is that the method is not reliable enough and certainly not consistent enough for absolute positioning. I can spend a whole week tweaking the code and the preset values and the pre-position values to get it to work. Two days later problems start happening again - it is very odd. This is with just one zoomie being tested. If I do test runs with a different Zoomie I have slightly different behaviours. Today (10/20/2024) will be my last day trying to find a universal solution to this problem and I will have to look at an alternative methodology. Thanks for the feedback and question, always great to hear from you! Cheers! Chris