Add FRC 2026 game piece detection models (#2332)

Added a model for detecting Rebuilt's FUEL game piece for versions of
the Orange Pi 5 and the Rubik Pi 3, which lowers the bar for less
experienced FRC teams to use object detection on their robots. Also
updated the documentation of object detection to reference the new
models.

The original model weights were trained by FRC team 2826 and were
published on this [Chief Delphi
thread](https://www.chiefdelphi.com/t/introducing-wave-robotics-yolov11-model-for-rebuilt/512701).

The dataset used for RKNN quantization is the [Fuel V3
dataset](https://universe.roboflow.com/frcroboraiders/frc-2026-fuel-sbrdk/dataset/3)
provided by
[frcroboraiders](https://universe.roboflow.com/frcroboraiders) on
Roboflow Universe.

docs/source/docs/objectDetection/about-object-detection.md

@@ -4,7 +4,7 @@
 
 PhotonVision supports object detection using neural network accelerator hardware, commonly known as an NPU. The two coprocessors currently supported are the {ref}`Orange Pi 5 <docs/objectDetection/opi:Orange Pi 5 (and variants) Object Detection>` and the {ref}`Rubik Pi 3 <docs/objectDetection/rubik:Rubik Pi 3 Object Detection>`.
 
-PhotonVision currently ships with a model trained on the [COCO dataset](https://cocodataset.org/) by [Ultralytics](https://github.com/ultralytics/ultralytics) (this model is licensed under [AGPLv3](https://www.gnu.org/licenses/agpl-3.0.en.html)). This model is meant to be used for testing and other miscellaneous purposes. It is not meant to be used in competition. For the 2025 post-season, PhotonVision also ships with a pretrained ALGAE model. A model to detect coral is available in the PhotonVision discord, but will not be distributed with PhotonVision.
+PhotonVision currently ships with a model trained on the [COCO dataset](https://cocodataset.org/) by [Ultralytics](https://github.com/ultralytics/ultralytics) (this model is licensed under [AGPLv3](https://www.gnu.org/licenses/agpl-3.0.en.html)). This model is meant to be used for testing and other miscellaneous purposes. It is not meant to be used in competition. For the 2026 season, PhotonVision ships with a model to detect FUEL, this is also licensed under AGPL.
 
 ## Tracking Objects
 

photon-core/src/main/java/org/photonvision/common/configuration/NeuralNetworkModelManager.java

@@ -149,16 +149,6 @@ public class NeuralNetworkModelManager {
                                 "hair drier", // Typo in official COCO documentation
                                 "toothbrush"));
 
-        nnProps.addModelProperties(
-                new ModelProperties(
-                        Path.of(modelsDirectory.getAbsolutePath(), "algaeV1-640-640-yolov8n.rknn"),
-                        "Algae v8n",
-                        new LinkedList<String>(List.of("Algae")),
-                        640,
-                        480,
-                        Family.RKNN,
-                        Version.YOLOV8));
-
         nnProps.addModelProperties(
                 new ModelProperties(
                         Path.of(modelsDirectory.getAbsolutePath(), "yolov8nCOCO.rknn"),
@@ -171,13 +161,13 @@ public class NeuralNetworkModelManager {
 
         nnProps.addModelProperties(
                 new ModelProperties(
-                        Path.of(modelsDirectory.getAbsolutePath(), "algae-coral-yolov8s.tflite"),
-                        "Algae Coral v8s",
-                        new LinkedList<String>(List.of("Algae", "Coral")),
+                        Path.of(modelsDirectory.getAbsolutePath(), "fuelV1-yolo11n.rknn"),
+                        "Fuel v11n",
+                        new LinkedList<String>(List.of("Fuel")),
                         640,
                         640,
-                        Family.RUBIK,
-                        Version.YOLOV8));
+                        Family.RKNN,
+                        Version.YOLOV11));
 
         nnProps.addModelProperties(
                 new ModelProperties(
@@ -189,6 +179,16 @@ public class NeuralNetworkModelManager {
                         Family.RUBIK,
                         Version.YOLOV8));
 
+        nnProps.addModelProperties(
+                new ModelProperties(
+                        Path.of(modelsDirectory.getAbsolutePath(), "fuelV1-yolo11n.tflite"),
+                        "Fuel v11n",
+                        new LinkedList<String>(List.of("Fuel")),
+                        640,
+                        640,
+                        Family.RUBIK,
+                        Version.YOLOV11));
+
         return nnProps;
     }