Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Small vision code cleanups/performance improvements #186

Merged
merged 2 commits into from
May 27, 2024
Merged

Small vision code cleanups/performance improvements #186

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7cdfe7b
vision: Use wpimath.objectToRobotPose in one-tag mode
auscompgeek Mar 10, 2024
b0f102a
vision: Remove redundant .radians() in Pose2d conversion
auscompgeek Mar 10, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 15 additions & 27 deletions components/vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@
from magicbot import tunable
from photonlibpy.photonCamera import PhotonCamera
from photonlibpy.photonTrackedTarget import PhotonTrackedTarget
from wpimath import objectToRobotPose
from wpimath.geometry import Pose2d, Rotation3d, Transform3d, Translation3d, Pose3d

from components.chassis import ChassisComponent
Expand Down Expand Up @@ -43,7 +44,8 @@ def __init__(
chassis: ChassisComponent,
) -> None:
self.camera = PhotonCamera(name)
self.camera_to_robot = Transform3d(pos, rot).inverse()
self.robot_to_camera = Transform3d(pos, rot)
self.camera_to_robot = self.robot_to_camera.inverse()
self.last_timestamp = -1
self.last_recieved_timestep = -1.0

Expand Down Expand Up @@ -92,26 +94,18 @@ def execute(self) -> None:

if self.should_log:
self.multi_best_log.setPose(
Pose2d(
p.best.x,
p.best.y,
p.best.rotation().toRotation2d().radians(),
)
Pose2d(p.best.x, p.best.y, p.best.rotation().toRotation2d())
)
self.multi_alt_log.setPose(
Pose2d(
p.alt.x,
p.alt.y,
p.alt.rotation().toRotation2d().radians(),
)
Pose2d(p.alt.x, p.alt.y, p.alt.rotation().toRotation2d())
)
else:
for target in results.getTargets():
# filter out likely bad targets
if target.getPoseAmbiguity() > 0.25:
continue

poses = estimate_poses_from_apriltag(self.camera_to_robot, target)
poses = estimate_poses_from_apriltag(self.robot_to_camera, target)
if poses is None:
# tag doesn't exist
continue
Expand All @@ -131,18 +125,14 @@ def execute(self) -> None:
Pose2d(
target.bestCameraToTarget.x,
target.bestCameraToTarget.y,
target.bestCameraToTarget.rotation()
.toRotation2d()
.radians(),
target.bestCameraToTarget.rotation().toRotation2d(),
)
)
self.single_alt_log.setPose(
Pose2d(
target.altCameraToTarget.x,
target.altCameraToTarget.y,
target.altCameraToTarget.rotation()
.toRotation2d()
.radians(),
target.altCameraToTarget.rotation().toRotation2d(),
)
)

Expand All @@ -151,22 +141,20 @@ def sees_target(self):


def estimate_poses_from_apriltag(
cam_to_robot: Transform3d, target: PhotonTrackedTarget
robot_to_camera: Transform3d, target: PhotonTrackedTarget
) -> Optional[tuple[Pose2d, Pose2d, float]]:
tag_id = target.getFiducialId()
tag_pose = apriltag_layout.getTagPose(tag_id)
if tag_pose is None:
return None

best_pose = tag_pose.transformBy(
target.getBestCameraToTarget().inverse()
).transformBy(cam_to_robot)
alternate_pose = (
tag_pose.transformBy(target.getAlternateCameraToTarget().inverse())
.transformBy(cam_to_robot)
.toPose2d()
best_pose = objectToRobotPose(
tag_pose, target.getBestCameraToTarget(), robot_to_camera
)
alternate_pose = objectToRobotPose(
tag_pose, target.getAlternateCameraToTarget(), robot_to_camera
)
return best_pose.toPose2d(), alternate_pose, best_pose.z
return best_pose.toPose2d(), alternate_pose.toPose2d(), best_pose.z


def get_target_skew(target: PhotonTrackedTarget) -> float:
Expand Down