Semi-Autonomous Rover Design for Exploration in Simulated Extraterrestrial Environments

The Shourya Rover is a semi-autonomous robotic platform designed for exploration in simulated extraterrestrial terrains. This paper presents a modular system architecture integrating high-performance computation, real-time sensing, and autonomous navigation. The rover employs an NVIDIA Jetson Orin Nano to process data from LiDAR, IMU, and depth cameras, supporting tasks such as obstacle detection, terrain mapping, and robotic manipulation. The control framework, developed in ROS Noetic, fuses multi-modal sensor data using an Extended Kalman Filter and implements path planning via the movebase stack. Extensive testing in Mars-analog environments demonstrated 25° slope climbing, 15 cm vertical obstacle traversal, and 0.66 m/s peak speed. Power and communication systems ensured continuous operation with minimal latency. The Shourya Rover serves as a scalable research platform for planetary exploration and autonomous system design.