# DDPG Robot Path Planning A beginner-friendly, single-file implementation of the Deep Deterministic Policy Gradient (DDPG) algorithm for continuous space robot path planning. ## Overview The script `ddpg_robot.py` trains a simple reinforcement learning agent to navigate a 2D continuous space to reach a goal while avoiding obstacles. The environment and the DDPG agent (both Actor and Critic networks) are contained entirely within this single script. ## Environment Details - **State Space**: 2D position $(x, y)$ - **Action Space**: 2D movement $(dx, dy)$ in continuous space - **Goal**: The agent is rewarded for minimizing the distance to the target goal `(5.0, 5.0)`. - **Max Steps**: 100 per episode. ## Prerequisites Be sure to install the required dependencies before running: ```bash pip install numpy torch ``` ## Usage Simply run the Python script: ```bash python ddpg_robot.py ``` The script will print out the cumulative reward for each episode. Over time (as observed in the 200 episodes), the agent will start stabilizing and learning the optimal path toward the goal!# Robot-Path-Planning-using-Deep-Deterministic-Policy-Gradient