Aerial-Ground Robotic system for Terrain estimation and Navigation
01 Jan 2019-
TL;DR: This work aims to build a base for developing in-house controllers and localization filters that can be used for real-time testing thereby decreasing the time for development.
Abstract: Unmanned ground vehicles and aerial vehicles can utilize their complementary features to address each other's inabilities like limited payload capacity, endurance, agility and field of view and thereby improving the mission efficiency. This paper presents an in-house developed ground and a micro aerial vehicle system performing a coordinated mission. The aerial vehicle is equipped with a stereo image sensor using which the environment around the ground vehicle is reconstructed in the form of an elevation map. This is then processed to get 2D traversability map based on which the ground vehicle performs path planning for a specified goal. The aerial robot also gives a sparse feature map which is used by the ground vehicle for initialization of its pose in the map. A pure pursuit controller for path following is implemented for both aerial and ground vehicle. Most of the proposed concepts are implemented and demonstrated on a real platform. This work also aims to build a base for developing in-house controllers and localization filters that can be used for real-time testing thereby decreasing the time for development.
Citations
More filters
Patent•
29 Oct 2020
TL;DR: In this paper, a storage device storing a movement assistance program is configured to cause a computer to implement the functions of: a terrain information acquisition instruction section (52) that instructs a terrain acquisition device (29) mounted on a flying machine (2) to acquire terrain information representing a terrain lying between a present location of a ground-moving machine (1) and a destination; a zone classification section (43) that makes a classification based on a predetermined movement permission condition and the terrain information to divide the terrain between the present location and the destination into a movement-permitted zone
Abstract: Provided is a storage device storing a movement assistance program, the program being configured to cause a computer to implement the functions of: a terrain information acquisition instruction section (52) that instructs a terrain information acquisition device (29) mounted on a flying machine (2) to acquire terrain information representing a terrain lying between a present location of a ground-moving machine (1) and a destination; a zone classification section (43) that makes a classification based on a predetermined movement permission condition and the terrain information to divide the terrain lying between the present location and the destination into a movement-permitted zone and a movement-prohibited zone; and an output section (44) that outputs at least one selected from a movement route of the ground-moving machine (1), the movement-permitted zone, and the movement-prohibited zone, the movement route leading from the present location to the destination without passing through the movement-prohibited zone.
References
More filters
Proceedings Article•
01 Jan 2009TL;DR: This paper discusses how ROS relates to existing robot software frameworks, and briefly overview some of the available application software which uses ROS.
Abstract: This paper gives an overview of ROS, an opensource robot operating system. ROS is not an operating system in the traditional sense of process management and scheduling; rather, it provides a structured communications layer above the host operating systems of a heterogenous compute cluster. In this paper, we discuss how ROS relates to existing robot software frameworks, and briefly overview some of the available application software which uses ROS.
8,387 citations
"Aerial-Ground Robotic system for Te..." refers background in this paper
...Through this work, we also aim to build a base for the in-house development of different controllers and localization filters that can be tested on a real platform with minimalistic changes by leveraging ROS capabilities [15]....
[...]
TL;DR: A survival of the fittest strategy that selects the points and keyframes of the reconstruction leads to excellent robustness and generates a compact and trackable map that only grows if the scene content changes, allowing lifelong operation.
Abstract: This paper presents ORB-SLAM, a feature-based monocular SLAM system that operates in real time, in small and large, indoor and outdoor environments. The system is robust to severe motion clutter, allows wide baseline loop closing and relocalization, and includes full automatic initialization. Building on excellent algorithms of recent years, we designed from scratch a novel system that uses the same features for all SLAM tasks: tracking, mapping, relocalization, and loop closing. A survival of the fittest strategy that selects the points and keyframes of the reconstruction leads to excellent robustness and generates a compact and trackable map that only grows if the scene content changes, allowing lifelong operation. We present an exhaustive evaluation in 27 sequences from the most popular datasets. ORB-SLAM achieves unprecedented performance with respect to other state-of-the-art monocular SLAM approaches. For the benefit of the community, we make the source code public.
3,807 citations
"Aerial-Ground Robotic system for Te..." refers methods in this paper
...When a loop is detected based on certain thresholds of matching the key points, a full map bundle adjustment(BA) is performed by the loop closure thread [9]....
[...]
TL;DR: ORB-SLAM2 as mentioned in this paper is a complete SLAM system for monocular, stereo and RGB-D cameras, including map reuse, loop closing and relocalization capabilities.
Abstract: We present ORB-SLAM2 a complete SLAM system for monocular, stereo and RGB-D cameras, including map reuse, loop closing and relocalization capabilities. The system works in real-time on standard CPUs in a wide variety of environments from small hand-held indoors sequences, to drones flying in industrial environments and cars driving around a city. Our back-end based on bundle adjustment with monocular and stereo observations allows for accurate trajectory estimation with metric scale. Our system includes a lightweight localization mode that leverages visual odometry tracks for unmapped regions and matches to map points that allow for zero-drift localization. The evaluation on 29 popular public sequences shows that our method achieves state-of-the-art accuracy, being in most cases the most accurate SLAM solution. We publish the source code, not only for the benefit of the SLAM community, but with the aim of being an out-of-the-box SLAM solution for researchers in other fields.
2,857 citations
TL;DR: An approach to robot perception and world modeling that uses a probabilistic tesselated representation of spatial information called the occupancy grid, a multidimensional random field that maintains stochastic estimates of the occupancy state of the cells in a spatial lattice is reviewed.
Abstract: An approach to robot perception and world modeling that uses a probabilistic tesselated representation of spatial information called the occupancy grid is reviewed. The occupancy grid is a multidimensional random field that maintains stochastic estimates of the occupancy state of the cells in a spatial lattice. To construct a sensor-derived map of the robot's world, the cell state estimates are obtained by interpreting the incoming range readings using probabilistic sensor models. Bayesian estimation procedures allow the incremental updating of the occupancy grid, using readings taken from several sensors over multiple points of view. The use of occupancy grids from mapping and for navigation is examined. Operations on occupancy grids and extensions of the occupancy grid framework are briefly considered. >
2,328 citations
01 Jan 2016
TL;DR: This chapter outlines how to integrate the grid map library into the reader's own applications and presents an application of the library for online elevation mapping with a legged robot.
Abstract: In this research chapter, we present our work on a universal grid map library for use as mapping framework for mobile robotics. It is designed for a wide range of applications such as online surface reconstruction and terrain interpretation for rough terrain navigation. Our software features multi-layered maps, computationally efficient repositioning of the map boundaries, and compatibility with existing ROS map message types. Data storage is based on the linear algebra library Eigen, offering a wide range of data processing algorithms. This chapter outlines how to integrate the grid map library into the reader’s own applications. We explain the concepts and provide code samples to discuss various features of the software. As a use case, we present an application of the library for online elevation mapping with a legged robot. The grid map library and the robot-centric elevation mapping framework are available open-source at http://github.com/ethz-asl/grid_map and http://github.com/ethz-asl/elevation_mapping.
198 citations
"Aerial-Ground Robotic system for Te..." refers methods in this paper
...The elevation mapping procedures developed by Peter et al in [12] and grid mapping framework developed in [13] are extensively used for dense reconstruction and traversability estimation....
[...]
...Elevation mapping is based on the grid maps ( also known as 2.5D maps) developed by the Peter et al [13] as they provide smooth handling of the multi-layered map....
[...]
...5D maps) developed by the Peter et al [13] as they provide smooth handling of the multi-layered map....
[...]
...The approaches of Peter et al, [7], Michael et al[8] are conceptually similar to the presented approach but they emphasize more on 3D LiDAR based dense reconstruction and gives a collaborative framework for legged and aerial robots....
[...]
...A package developed by Peter et al [12] is used to generate the elevation map of the terrain....
[...]