Adaptive Hybrid Navigation for Autonomous Mobile Robots in GPS-Denied Environments

Track

Navigation — Topic 02

Project description

Autonomous mobile robots operating in outdoor environments often rely on GPS/GNSS for global localization and navigation. However, satellite signals can become unavailable or unreliable in certain environments, making continuous autonomous navigation challenging.

This project focuses on developing a hybrid navigation system capable of operating with and without GPS availability. The system should automatically adapt to changes in signal availability by combining satellite-based positioning with onboard localization techniques, ensuring continuous navigation and reliable position estimation even when GPS signals are lost.

Objective

Develop an adaptive navigation system that enables a mobile robot to navigate seamlessly between GPS-assisted and GPS-denied operating modes. The approach will combine GPS/GNSS data with LiDAR, IMU and camera-based localization techniques through sensor fusion.

The system should:

• Detect GPS signal degradation or loss
• Automatically transition to an alternative localization strategy
• Maintain accurate positioning and autonomous navigation until GPS availability is restored

Required skills

• Proficiency in Python and C++
• Experience with ROS 2
• Knowledge of autonomous navigation and localization
• Understanding of sensor fusion and coordinate transformations
• Familiarity with GPS/GNSS, LiDAR, IMU and camera-based perception
• Knowledge of SLAM, visual odometry and LiDAR odometry
• Experience with Nav2 and localization algorithms
• Understanding of Kalman Filters, including EKF/UKF

Recommended skills

• Familiarity with Gazebo or other robotics simulation environments