An Autonomously Navigation System for Forestry Quadrotor within GPS-denied Below-canopy Environment

This paper presents a navigation system that enables quadrotor to navigate autonomously within GPS-denied below-canopy environment without GPS. The navigation system consists of the function of Simultaneous Localization and Mapping(SLAM) and path planning. The online real-time locating estimation is achieved by Semi-direct Visual Odometer(SVO) to calculate the odometer data; we locate the position using the fusion of laser radar data and the visual odometer in the front end, the graph-based optimization is used as an optimization mechanism to map in the back end. The path planning finds an obstacle-free trajectory based on the waypoints generated by the improved RRT^* algorithm fitted with the Bezier curve.