update documentation

doc/general/architecture.md

@@ -3,24 +3,24 @@
 **Summary:** This page gives an overview over the planned general architecture of the vehicle agent.
 The document contains an overview over all [nodes](#overview) and [topics](#topics).
 
-- [Planned architecture of vehicle agent](#planned-architecture-of-vehicle-agent)
-  - [Overview](#overview)
-  - [Perception](#perception)
-    - [Obstacle Detection and Classification](#obstacle-detection-and-classification)
-    - [Traffic Light Detection](#traffic-light-detection)
-    - [Localization](#localization)
-  - [Planning](#planning)
-    - [Global Planning](#global-planning)
-    - [Decision Making](#decision-making)
-    - [Local Planning](#local-planning)
-      - [Collision Check](#collision-check)
-      - [ACC](#acc)
-      - [Motion Planning](#motion-planning)
-  - [Acting](#acting)
-    - [Path following with Steering Controllers](#path-following-with-steering-controllers)
-    - [Velocity control](#velocity-control)
-    - [Vehicle controller](#vehicle-controller)
-  - [Visualization](#visualization)
+- [Overview](#overview)
+- [Perception](#perception)
+  - [Obstacle Detection and Classification](#obstacle-detection-and-classification)
+  - [Traffic Light Detection](#traffic-light-detection)
+  - [Localization](#localization)
+  - [Lane Detection](#lane-detection)
+- [Planning](#planning)
+  - [Global Planning](#global-planning)
+  - [Decision Making](#decision-making)
+  - [Local Planning](#local-planning)
+    - [Collision Check](#collision-check)
+    - [ACC](#acc)
+    - [Motion Planning](#motion-planning)
+- [Acting](#acting)
+  - [Path following with Steering Controllers](#path-following-with-steering-controllers)
+  - [Velocity control](#velocity-control)
+  - [Vehicle controller](#vehicle-controller)
+- [Visualization](#visualization)
 
 ## Overview
 
@@ -99,6 +99,18 @@ Publishes:
 
 - ```ego_position``` ([nav_msgs/Odometry Message](http://docs.ros.org/en/noetic/api/nav_msgs/html/msg/Odometry.html))
 
+### Lane Detection
+
+Provides a mask and an image where the lane is drawn in.
+
+Subscriptions:
+
+- ```rgb_camera``` ([sensor_msgs/Image](https://docs.ros.org/en/api/sensor_msgs/html/msg/Image.html))
+
+Publishes:
+
+- ```lane_detection_mask``` ([sensor_msgs/Image](https://docs.ros.org/en/api/sensor_msgs/html/msg/Image.html))
+- ```lane_detection_image``` ([sensor_msgs/Image](https://docs.ros.org/en/api/sensor_msgs/html/msg/Image.html))
 ## Planning
 
 The planning uses the data from the [Perception](#Perception) to find a path on which the ego vehicle can safely reach

doc/perception/lane_detection_node.md

@@ -0,0 +1,36 @@
+# Lane Detection
+
+**Summary:** This page gives an overview over the implementation of the lane detection.
+
+- [Input](#input)
+- [Detection](#detection)
+  - [OpenCV Model](#opencv-model)
+    - [Model](#model)
+    - [Outputs](#outputs)
+
+
+## Input
+
+The input is the image of the "center" camera, which is used to detect the lanes.
+
+## Detection
+
+### OpenCV Model
+
+#### Model
+
+The lane detection algorithm involves preprocessing the input image to enhance lane visibility, converting it to grayscale, and applying edge detection (Canny edge detection).
+Next, it isolates the region of interest, by masking the lane area using a trapezoidal shape.
+The algorithm then uses the Hough Line Transform to detect lines corresponding to the lanes and filters them based on their slopes and positions. Finally, it overlays the detected lanes on the original image.
+
+![OpenCV Lane Detection flow](../assets/perception/OpenCV_lane_detection_flow.png)
+
+#### Outputs
+
+The mask contains the left and right line of the lane. The mask is an array of type uint8
+
+![OpenCV Lane Detection Mask](../assets/perception/OpenCV_Lane_Detection_Mask.png)
+
+This model also outputs an image with already drawn in lines
+
+![OpenCV Lane Detection Image](../assets/perception/OpenCV_Lane_Detection_Image.png)