Kyle Stachowicz

The Geometry2d library's hitSet currently incurs a _lot_ of overhead by actually creating new smart pointers. We could save a lot of allocation and reference counting by just returning an...

[TODO] Make trapezoidal motion work correctly when initial velocity > maximum velocity

2

Currently, if the initial velocity is greater than the maximum velocity, trapezoidal motion will give discontinuous paths. This case should be handling by decelerating to maximum speed, moving at maximum...

https://github.com/RoboJackets/robocup-software/blob/3a4185a6767fb4d1c507711b091c58f815d752fe/soccer/motion/MotionControl.cpp#L106 From review of #1494. This value should be stored in `RobotIntent`.

Pull request: #1495. Link to TODO, or file number/line: Processor.cpp:251

As of #1494 we have a significant amount of redundancy between the implementation of `TrajectoryHitStatic` and `TrajectoryHitsDynamic`. Ideally we should abstract out the shared parts. https://github.com/RoboJackets/robocup-software/pull/1494/files/1d6b526aaeada2dc275fe30c3e2316de4e177dd9#diff-24812fbd3fbc6c0a2298ba50adc23eb4R7 Brought up in review...

Currently these functions give slightly (`O(1e-3)`) different results. We should investigate this and make sure it's not a bug somewhere in the implementation. Brought up in review for #1493.

I'd like to change the geometry of a HField at runtime. I am using the following code: ```python hfield = physics.bind(self._hfield_chunks[geom_idx]) res = hfield.nrow # Generate terrain in a 2D...

Experimental and untested but if we can support the bandwidth and memory (spoiler: we almost certainly can) this should make complex data interchange and robot configuration a lot easier. This...

We need to port the radio to RTOS so we can take advantage of yielding (instead of just delaying for 1ms). This should basically look like the following: ```cpp static...