Online Aerial Terrain Mapping for Ground Robot Navigation

Abstract

This work presents a collaborative unmanned aerial and ground vehicle system which utilizes the aerial vehicle’s overhead view to inform the ground vehicle’s path planning in real time. The aerial vehicle acquires imagery which is assembled into a orthomosaic and then classified. These terrain classes are used to estimate relative navigation costs for the ground vehicle so energy-efficient paths may be generated and then executed. The two vehicles are registered in a common coordinate frame using a real-time kinematic global positioning system (RTK GPS) and all image processing is performed onboard the unmanned aerial vehicle, which minimizes the data exchanged between the vehicles. This paper describes the architecture of the system and quantifies the registration errors between the vehicles.

Experiments in unmanned aerial vehicle/unmanned ground vehicle radiation search

“…The custom multirotor, shown in Figure , was originally developed by a Virginia Tech senior design team for a precision package delivery competition they were tasked to create with NASA Langley funding. This multirotor was adapter for use in Peterson et al () and further altered for this project. This UAV utilizes a Pixahawk flight controller running the Arducopter firmware and uses an onboard Nvidia TX2 to run the high‐level ROS‐based software system.…”

“…The following sections will describe the hardware and algorithms used in the experiment conducted at SRNL. The system presented here is an evolution of the system described in Peterson, Chaudhry, Abdelatty, Bird, and Kochersberger (2018). Like the earlier version, this system uses the robot operating system (ROS; Quigley et al, 2009) to integrate the various software components.…”

“…In a separate project, described in part in Peterson et al (2018), we are working on an online 3D reconstruction method, that would largely address both the issue of registering the map coordinate frame with the base station as well as the time required to generate the reconstruction. The aim of this other work is to enable the operation of the multirobot system without complete knowledge of the geometry of the environment, and to explore the environment with a UAV to enable a blind UGV to navigate safely.…”

A Quadral-Fuzzy Control Approach to Flight Formation by a Fleet of Unmanned Aerial Vehicles

“…However, several challenges must be solved for the service to be effective [6]- [8]: (1) Limited payload: in general, goods must not weigh more than 2 kg; (2) Integration of low-cost sensors and positioning system, that is, several sensors like gyroscope, accelerometer, among others, can be used to create the odometry. The sensor fusion with accurate high location sensors, such as Real Time Kinematic GPS [9], allows to obtain the drone position in a global reference system; (3) Avoid obstacles and collisions: it is necessary to establish a flyable collision-free path in a dynamic environment; (4) Communication and connectivity: communication links with the ground control station are need to receive instructions; (5) Landing at specified locations or the use of a parachute to delivery goods, (6) Limited flight range due to energy requirement (battery duration): the traveled distance depends on power transfer efficiency for motor, cruising velocity and power consumption of electronics, (7) Other concerning including government regulation and public acceptation.…”

What Lies Beneath One’s Feet? Terrain Classification Using Inertial Data of Human Walk

“…Similarly, Ma et al employed aerial image data for terrain classification to support off-road navigation of the ground vehicle using low-rank sparse representation. Peterson et al [12] also used imagery data using aerial vehicle for ground robot navigation. With the aim of terrain classification, Dornik et al [6] classified different soil types using geographic object-based analysis on images.…”