Merge pull request #950 from Panchadip-128/Added-Drone-navigation-Det…

…ection-using-RL

Drone Navigation Detection and Path Tracing

Drone Navigation Detection using Reinforcement Learning techniques/Dataset/Dataset- Explanations

@@ -0,0 +1,53 @@
+Detailed Description of the Dataset for Drone Navigation Project
+1. Environment State Representation
+The environment for the drone navigation task is modeled as a 2D grid (10x10) where each cell can represent different types of entities that the drone interacts with. The key components are:
+
+Free Space: This represents areas of the grid where the drone can move freely. Free space cells are the navigable areas where the drone does not encounter any obstacles.
+
+Obstacles: These are fixed points on the grid that the drone must avoid to prevent collisions. In this project, obstacles are defined as specific coordinates:
+
+Example:
+Obstacle 1: (6, 6)
+Obstacle 2: (7, 7)
+Target: This is the desired destination that the drone aims to reach. The target position is critical for the navigation algorithm to determine successful completion of the task.
+
+Example:
+Target Position: (8, 8)
+2. State Space
+The state of the drone is represented using a 2D NumPy array with two elements, denoting the drone's current position on the grid:
+
+state[0]: Represents the x-coordinate (horizontal position) of the drone.
+state[1]: Represents the y-coordinate (vertical position) of the drone.
+The observation space is defined within the bounds of the grid, specifically from 0 to 10. This range indicates that the drone's movements and positions are confined within a 10x10 grid.
+
+3. Action Space
+The available actions for the drone are discrete movements within the grid. Each action corresponds to a direction the drone can move:
+
+0: Up (increases y-coordinate)
+1: Down (decreases y-coordinate)
+2: Left (decreases x-coordinate)
+3: Right (increases x-coordinate)
+4: Up-Right (increases both x and y coordinates)
+5: Up-Left (decreases x and increases y coordinates)
+6: Down-Right (increases x and decreases y coordinates)
+7: Down-Left (decreases both x and y coordinates)
+This action space allows for basic directional movements, enabling the drone to navigate towards its target while avoiding obstacles.
+
+4. Sample Data
+While the environment is not reliant on external datasets, the positions of obstacles and the target can be treated as parameters that define the specific scenario of the navigation task. Below are examples of the parameters used in the project:
+
+Initial State: The drone starts at position (5, 5).
+Obstacles: [(6, 6), (7, 7)]
+Target Position: (8, 8)
+This setup allows for a controlled testing environment where various navigation strategies can be implemented and evaluated.
+
+5. Data Generation
+The grid layout, positions of obstacles, and the target location are configurable parameters that can be adjusted to create different scenarios for testing the drone's navigation algorithm. The drone can be tested in various configurations to analyze its performance in navigating towards the target while avoiding collisions.
+
+6. Future Dataset Enhancements
+In future iterations of this project, there are several potential enhancements that can be made to the dataset:
+
+Dynamic Obstacles: Introducing moving obstacles that change positions over time, simulating more realistic navigation challenges.
+Variable Target Locations: Allowing the target position to change during the task to test the drone's adaptability and decision-making.
+Real-World Data: Integrating real-world datasets (such as GPS coordinates or aerial maps) to enhance the environment's complexity and realism.
+Multiple Drones: Expanding the project to include multiple drones navigating the same environment, which could lead to more complex scenarios and interactions.

Drone Navigation Detection using Reinforcement Learning techniques/Dataset/datset_info

@@ -0,0 +1,37 @@
+# Dataset Description for Drone Navigation Project
+
+## Environment State Representation:
+The drone's environment is represented as a 2D grid where each cell can represent different entities:
+- **Free Space**: Areas where the drone can navigate.
+- **Obstacles**: Fixed points on the grid that the drone must avoid.
+  - Example: (6, 6), (7, 7)
+- **Target**: The desired destination for the drone to reach.
+  - Example: (8, 8)
+
+## State Space:
+The state of the drone is represented as a 2D NumPy array with two elements:
+- **state[0]**: The x-coordinate of the drone's current position.
+- **state[1]**: The y-coordinate of the drone's current position.
+
+The observation space is defined within the bounds of the grid, specifically [0, 10], indicating that the drone can move within a 10x10 grid.
+
+## Action Space:
+The actions available to the drone are represented as discrete movements within the grid:
+- **0**: Up
+- **1**: Down
+- **2**: Left
+- **3**: Right
+- **4**: Up-Right
+- **5**: Up-Left
+- **6**: Down-Right
+- **7**: Down-Left
+
+## Sample Data:
+While the environment does not rely on external datasets, the positions of obstacles and the target can be seen as parameters that define the specific scenario of the navigation task.
+
+## Data Generation:
+The grid layout, obstacle positions, and target location can be adjusted as necessary to create various scenarios for testing the drone's navigation algorithm.
+
+## Future Dataset Enhancements:
+Future versions of the project may incorporate more complex environments with variable obstacle positions, dynamic targets, and real-world data, enhancing the robustness and adaptability of the navigation algorithm.
+

Drone Navigation Detection using Reinforcement Learning techniques/Images/img files

@@ -0,0 +1 @@
+