Autonomy Bootcamp DONE

modules/bootcamp/decision_simple_waypoint.py

@@ -36,8 +36,8 @@ def __init__(self, waypoint: location.Location, acceptance_radius: float) -> Non
         # ============
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
-
         # Add your own
+        self.has_landed = False
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
@@ -67,11 +67,26 @@ def run(
         # ============
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
+        # Calculate distance to waypoint
+        delta_x = self.waypoint.location_x - report.position.location_x
+        delta_y = self.waypoint.location_y - report.position.location_y
+        dist = (delta_x**2 + delta_y**2) ** 0.5
+
+        if self.has_landed:
+            return command
 
-        # Do something based on the report and the state of this class...
+        if report.status.name == "LANDED":
+            self.has_landed = True
+            return command
+
+        if dist > self.acceptance_radius:
+            if report.status.name == "HALTED":
+                command = commands.Command.create_set_relative_destination_command(delta_x, delta_y)
+        else:
+            if report.status.name == "HALTED":
+                command = commands.Command.create_land_command()
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
         # ============
-
         return command

modules/bootcamp/decision_waypoint_landing_pads.py

@@ -36,8 +36,9 @@ def __init__(self, waypoint: location.Location, acceptance_radius: float) -> Non
         # ============
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
-
-        # Add your own
+        self.has_landed = False
+        self.reached_waypoint = False
+        self.closest_landing_pad = None
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
@@ -67,11 +68,47 @@ def run(
         # ============
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
-
-        # Do something based on the report and the state of this class...
+        if self.has_landed:
+            return command
+
+        if report.status.name == "LANDED":
+            self.has_landed = True
+            return command
+
+        if not self.reached_waypoint:
+            # Calculate distance to waypoint
+            delta_x = self.waypoint.location_x - report.position.location_x
+            delta_y = self.waypoint.location_y - report.position.location_y
+            dist = (delta_x**2 + delta_y**2) ** 0.5
+
+            if dist > self.acceptance_radius:
+                if report.status.name == "HALTED":
+                    command = commands.Command.create_set_relative_destination_command(delta_x, delta_y)
+            else:
+                self.reached_waypoint = True
+                # Find closest landing pad
+                min_dist = float("inf")
+                for pad in landing_pad_locations:
+                    pad_delta_x = pad.location_x - report.position.location_x
+                    pad_delta_y = pad.location_y - report.position.location_y
+                    pad_dist = (pad_delta_x**2 + pad_delta_y**2) ** 0.5
+                    if pad_dist < min_dist:
+                        min_dist = pad_dist
+                        self.closest_landing_pad = pad
+        else:
+            # Move to closest landing pad
+            delta_x = self.closest_landing_pad.location_x - report.position.location_x
+            delta_y = self.closest_landing_pad.location_y - report.position.location_y
+            dist = (delta_x**2 + delta_y**2) ** 0.5
+
+            if dist > self.acceptance_radius:
+                if report.status.name == "HALTED":
+                    command = commands.Command.create_set_relative_destination_command(delta_x, delta_y)
+            else:
+                if report.status.name == "HALTED":
+                    command = commands.Command.create_land_command()
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
         # ============
-
         return command

modules/bootcamp/detect_landing_pad.py

@@ -86,43 +86,40 @@ def run(self, image: np.ndarray) -> "tuple[list[bounding_box.BoundingBox], np.nd
         Return: A tuple of (list of bounding boxes, annotated image) .
             The list of bounding boxes can be empty.
         """
-        # ============
-        # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
-        # ============
-
-        # Ultralytics has documentation and examples
-
-        # Use the model's predict() method to run inference
-        # Parameters of interest:
-        # * source
-        # * conf
-        # * device
-        # * verbose
-        predictions = ...
+        # Run inference using the model's predict method
+        # Set confidence threshold to 0.7, use the correct device, and suppress verbose output
+        predictions = self.__model.predict(
+            source=image,
+            conf=0.7,
+            device=self.__DEVICE,
+            verbose=False,
+        )
 
-        # Get the Result object
-        prediction = ...
+        # Get the first (and only) Result object
+        prediction = predictions[0]
 
-        # Plot the annotated image from the Result object
-        # Include the confidence value
-        image_annotated = ...
+        # Plot the annotated image from the Result object (includes confidence values)
+        image_annotated = prediction.plot()
 
         # Get the xyxy boxes list from the Boxes object in the Result object
-        boxes_xyxy = ...
-
-        # Detach the xyxy boxes to make a copy,
-        # move the copy into CPU space,
-        # and convert to a numpy array
-        boxes_cpu = ...
+        boxes = prediction.boxes
+        if boxes is not None and boxes.xyxy is not None:
+            boxes_xyxy = boxes.xyxy
+        else:
+            boxes_xyxy = []
+
+        # Detach the xyxy boxes to make a copy, move to CPU, and convert to numpy array
+        if hasattr(boxes_xyxy, "cpu"):
+            boxes_cpu = boxes_xyxy.cpu().numpy()
+        else:
+            boxes_cpu = np.array(boxes_xyxy)
 
         # Loop over the boxes list and create a list of bounding boxes
         bounding_boxes = []
-        # Hint: .shape gets the dimensions of the numpy array
-        # for i in range(0, ...):
-        #     # Create BoundingBox object and append to list
-        #     result, box = ...
-
-        return [], image_annotated
-        # ============
-        # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
-        # ============
+        for i in range(boxes_cpu.shape[0]):
+            bounds = boxes_cpu[i]
+            result, box = bounding_box.BoundingBox.create(bounds)
+            if result and box is not None:
+                bounding_boxes.append(box)
+
+        return bounding_boxes, image_annotated