Fulfilling the final project requirements for Stanford's AA274a Principles of Robot Autonomy I course, a fully autonomous search and rescue mission was carried out in Gazebo.  The robot is tasked with locating all objects in the maze, then navigating to desired ones to "rescue" them. The robot then returns to the start.

This project implements and relies upon autonomous frontier exploration, A-star path planning coupled with real trajectory controllers, EKF SLAM, and a convolutional neural network for object detection. The work was carried out in ROS and rospy.