OP late braking for lead (planner lag and actuator lag)
Describe the bug
Yesterday I had my first instance of OP not braking in time for a lead car and triggering the AEB warning: I felt like I had to slam on the brakes to stop safely. I'm on stock 0.8.15 on a Comma 3 with a 2018 RAV4 Hybrid + SDSU.
Provide a route where the issue occurs
b4069565beef1091|2022-07-26--17-20-36--5
openpilot version
0.8.15
Additional info
No response
The lead car braked pretty hard (-5 m/s^2), but looks like the car was slow in responding to the ACC command. Probably a duplicate of https://github.com/commaai/openpilot/issues/2731.
@sshane any ideas why it didn't respond?

Interesting, thanks for the plot :) I was trying to get plotjuggler going, but I'm still on Ubuntu 18.04, so no dice. Note that my RAV4 should be on TSS-P, not TSS2, although I guess the Hybrid model is a bit weird in that regard?
While the car did respond late to the requests late (and I'm not totally sure why), I also see the planner was pretty delayed to request enough braking. For example, while the lead is at -2.3 m/s/s we're only requesting -0.8 m/s/s

Is there anything else I can do to help diagnose this? Or are you going to put this item on your list and work on it internally?
Adding 2 similar issues here https://github.com/commaai/openpilot/issues/20970 and https://github.com/commaai/openpilot/issues/25194
Master has improved Toyota longitudinal now.