My team and I are looking to improve our autonomous skills, and were thinking of trying to track the trajectory of the peojectile triballs as the leave the flywheel with a grid of ultrasounic sensors, then using previously collected data points and a probability algorithm to predict where each triball will most likely land, and find the most effecient route every run through. Does this sound possible? Im unsure wether the brain and code will enable this level of tracking
I think this is something that you could definitely try and utilize. Knowing the spatial information of the triball projectiles as you launch them can be used to calculate where the triballs will land, but the movement of the triballs after they land could affect the accuracy of this information. Depending on the design of your launcher, the triball could roll around a lot from how it originally landed, so the route found may not be the most effective. I’m unsure if the distance sensor or the older ultrasonic sensor could record values fast enough to determine a velocity, but I think it is possible.
If your launcher could consistently get triballs in a certain area near the goal, I think this can help determine a path for the robot to follow. Even if you do, I don’t believe an optimized route would significantly differ from a ‘one-size-fits-all’ pre-planned solution for the average robot. Using a vision sensor or distance sensor on the robot could potentially be better at determining triball locations on the field, in my opinion, if you think that this information would still be helpful.
1 Like
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.