Initial Submission - Manjot

modules/bootcamp/decision_simple_waypoint.py

@@ -37,7 +37,8 @@ def __init__(self, waypoint: location.Location, acceptance_radius: float) -> Non
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
 
-        # Add your own
+        # Boolean to check for completion
+        self.waypoint_reached = False
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
@@ -68,7 +69,23 @@ def run(
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
 
-        # Do something based on the report and the state of this class...
+        # Check the report to see what the status of the drone is
+        if report.status == drone_status.DroneStatus.HALTED:
+
+            # First check whether it has gotten to the waypoint
+            if not self.waypoint_reached:
+                # Moves towards the waypoint
+                x_dist = self.waypoint.location_x - report.position.location_x
+                y_dist = self.waypoint.location_y - report.position.location_y
+                command = commands.Command.create_set_relative_destination_command(x_dist, y_dist)
+
+                self.waypoint_reached = True
+
+            # Once it has gotten to waypoint, land it
+            else:
+                command = commands.Command.create_land_command()
+
+        # raise NotImplementedError
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑

modules/bootcamp/decision_waypoint_landing_pads.py

@@ -36,9 +36,8 @@ def __init__(self, waypoint: location.Location, acceptance_radius: float) -> Non
         # ============
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
-
-        # Add your own
-
+        self.landpad_reached = False
+        self.waypoint_reached = False
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
         # ============
@@ -68,10 +67,54 @@ def run(
         # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
         # ============
 
-        # Do something based on the report and the state of this class...
+        if report.status == drone_status.DroneStatus.HALTED:
+
+            if not self.waypoint_reached:
+                # Moves towards the waypoint
+                x_dist = self.waypoint.location_x - report.position.location_x
+                y_dist = self.waypoint.location_y - report.position.location_y
+                command = commands.Command.create_set_relative_destination_command(x_dist, y_dist)
+
+                self.waypoint_reached = True
+
+            # Once you get to the waypoint, you have to go to the nearest landpad
+            elif not self.landpad_reached:
+                # Find the closest pad and store it in closest_landpad
+                #    This is using the min() method to iterate through each pad, get the distance to it and compare it with others
+                closest_pad = min(
+                    landing_pad_locations, key=lambda pad: self.get_distance(report.position, pad)
+                )
+                print(f"Closest Landpad: {closest_pad.location_x}, {closest_pad.location_y}")
+
+                # Move to the closest landpad
+                x_dist = closest_pad.location_x - report.position.location_x
+                y_dist = closest_pad.location_y - report.position.location_y
+                # print(f"{x_dist}, {y_dist}")
+                command = commands.Command.create_set_relative_destination_command(x_dist, y_dist)
+
+                self.landpad_reached = True
+
+            else:
+                command = commands.Command.create_land_command()
+
+        # raise NotImplementedError
+
+        return command
+
+    # Method used to check distance sqaured between two locations
+    #   We don't need the square root since this is for comparison only
+    def get_distance(self, loc_1: location.Location, loc_2: location.Location) -> float:
+        """
+        This method is used to get the distance squared between any two locations
+
+        """
+        x = loc_1.location_x - loc_2.location_x
+        y = loc_1.location_y - loc_2.location_y
+
+        # Don't need to square root when comparing between squares
+        hypotenuse_squared = (x**2) + (y**2)
+        return hypotenuse_squared
 
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
         # ============
-
-        return command

modules/bootcamp/detect_landing_pad.py

@@ -16,10 +16,7 @@
 # ============
 # ↓ BOOTCAMPERS MODIFY BELOW THIS COMMENT ↓
 # ============
-# Bootcampers remove the following lines:
-# Allow linters and formatters to pass for bootcamp maintainers
-# No enable
-# pylint: disable=unused-argument,unused-private-member,unused-variable
+
 # ============
 # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
 # ============
@@ -98,31 +95,41 @@ def run(self, image: np.ndarray) -> "tuple[list[bounding_box.BoundingBox], np.nd
         # * conf
         # * device
         # * verbose
-        predictions = ...
+        predictions = self.__model.predict(
+            source=image, conf=0.7, device=self.__DEVICE, verbose=False
+        )
 
         # Get the Result object
-        prediction = ...
-
+        # Only 1 image was used therefore we access the first result in the list
+        prediction = predictions[0]
         # Plot the annotated image from the Result object
         # Include the confidence value
-        image_annotated = ...
+        image_annotated = prediction.plot(conf=True, boxes=True)
 
         # Get the xyxy boxes list from the Boxes object in the Result object
-        boxes_xyxy = ...
+        boxes_xyxy = prediction.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_cpu = boxes_xyxy.detach().cpu().numpy()
 
         # Loop over the boxes list and create a list of bounding boxes
         bounding_boxes = []
+        # .shape returns a tuple with the 0th index being the number of lists in the numpy array
+        for i in range(0, boxes_cpu.shape[0]):
+            # This func returns a tuple, with whether the operation was a sucess (bool) & the bounding box itself
+            successful, box = bounding_box.BoundingBox.create(bounds=boxes_cpu[i])
+
+            if successful:
+                bounding_boxes.append(box)
+
         # 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
+        return bounding_boxes, image_annotated
         # ============
         # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑
         # ============

modules/bootcamp/tests/run_decision_example.py

@@ -21,7 +21,7 @@
 # to reach the 1st command
 # Increase the step size if your computer is lagging
 # Larger step size is smaller FPS
-TIME_STEP_SIZE = 0.1  # seconds
+TIME_STEP_SIZE = 0.15  # seconds
 
 # OpenCV ignores your display settings,
 # so if the window is too small or too large,

requirements.txt

@@ -17,7 +17,7 @@ pytest
 # If you are on MacOS, don't have a CUDA capable GPU,
 # or if things don't work and you just want to use the CPU,
 # comment out the "--extra-index-url" option
---extra-index-url https://download.pytorch.org/whl/cu124
+# --extra-index-url https://download.pytorch.org/whl/cu124
 
 # ============
 # ↑ BOOTCAMPERS MODIFY ABOVE THIS COMMENT ↑