Real-time autonomous 3D navigation for tracked vehicles in rescue environments

Menna, Matteo; Gianni, Mario; Ferri, F.; Pirri, Fiora

doi:10.1109/iros.2014.6942634

Cited by 26 publications

(34 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, segmentation is performed and clusters are labeled according to geometrical constraints applied to surface normal directions, mean curvatures and 3D-coordinates of points. This results in a classification of the environment in regions such as walls, terrain, surmountable obstacles and stairs/ramps [4,5]. Traversability is then computed as a cost function taking into account the point cloud classification and the local geometric features [4,5] (such as obstacle clearance, terrain roughness and point cloud density).…”

Section: B Path Planning On Built Mapsmentioning

confidence: 99%

“…On both global and local scale, the connectivity of the traversable terrain is captured by using a sampling-based approach. In particular, a tree is expanded in the configuration space by using a randomized A* approach [4,5].…”

Section: B Path Planning On Built Mapsmentioning

confidence: 99%

“…The 3D path planning and navigation methods presented in this paper are based on the works [4,5,6]. The underlying modules provide functionalities such as real-time point cloud segmentation and traversability analysis.…”

Section: Introductionmentioning

confidence: 99%

“…While providing precise local reconstruction of an environment, this technique can not improve the map in the event of place recognition. The localization algorithm presented in this paper is therefore based on the pose-graph SLAM strategy as described in [3].The 3D path planning and navigation methods presented in this paper are based on the works [4,5,6]. The underlying modules provide functionalities such as real-time point cloud segmentation and traversability analysis.…”

mentioning

confidence: 99%

See 3 more Smart Citations

3D localization, mapping and path planning for search and rescue operations

Dubé

Gawel

Cadena

et al. 2016

2016 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)

View full text Add to dashboard Cite

Abstract-This work presents our results on 3D robot localization, mapping and path planning for the latest joint exercise of the European project "Long-Term Human-Robot Teaming for Robots Assisted Disaster Response" (TRADR)1 . The full system is operated and evaluated by firemen end-users in realworld search and rescue experiments. We demonstrate that the system is able to plan a path to a goal position desired by the fireman operator in the TRADR Operational Control Unit (OCU), using a persistent 3D map created by the robot during previous sorties. I. INTRODUCTIONTeams of autonomous mobile robots have the potential to reduce human risks during disaster response as well as the associated costs [1]. Different levels of robot autonomy are required in order to effectively support a rescue squad performing high-level tasks such as exploring the disaster area, detecting victims and taking chemical samples. Moreover, long-term operation of robotic platforms is desired for humans and robots to collaborate over several days of disaster intervention. To this end, building and maintaining a persistent representation of the environment, accurate localization, and efficient path planning are fundamental prerequisites.Prior to this work, a SLAM strategy based on Iterative Closest Point (ICP) for the robotic platform considered in this work was proposed in [2]. While providing precise local reconstruction of an environment, this technique can not improve the map in the event of place recognition. The localization algorithm presented in this paper is therefore based on the pose-graph SLAM strategy as described in [3].The 3D path planning and navigation methods presented in this paper are based on the works [4,5,6]. The underlying modules provide functionalities such as real-time point cloud segmentation and traversability analysis. A randomized A* approach is applied on the current terrain structure interpretation.In the remainder of this report, we concisely describe the localization, mapping and path planning systems and present the results of experiments with firemen end-users, at the latest TRADR Joint Exercise (TJEx).II. SYSTEM DESCRIPTION While the TRADR system comprises an integrated framework spanning from low-level perception functionalities to high-level reasoning, in this work we focus on presenting the latest advances in the integrated SLAM and path planning.

show abstract

Section: B Path Planning On Built Mapsmentioning

confidence: 99%

Section: B Path Planning On Built Mapsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

3D localization, mapping and path planning for search and rescue operations

Dubé

Gawel

Cadena

et al. 2016

2016 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)

View full text Add to dashboard Cite

show abstract

“…Three-dimensional (3D) lidar sensors are a key technology for navigation, localization, mapping and scene understanding in novel ground vehicle systems such as autonomous cars [1], search and rescue robots [2], and planetary exploration rovers [3]. One major limitation regarding the use of lidar technology in these challenging applications is the time and computational resources required to process the dense point clouds generated by these sensors.…”

Section: Introductionmentioning

confidence: 99%

Voxel-Based Neighborhood for Spatial Shape Pattern Classification of Lidar Point Clouds with Supervised Learning

Plaza-Leiva

Gómez-Ruiz

Mandow

et al. 2017

Sensors

View full text Add to dashboard Cite

Improving the effectiveness of spatial shape features classification from 3D lidar data is very relevant because it is largely used as a fundamental step towards higher level scene understanding challenges of autonomous vehicles and terrestrial robots. In this sense, computing neighborhood for points in dense scans becomes a costly process for both training and classification. This paper proposes a new general framework for implementing and comparing different supervised learning classifiers with a simple voxel-based neighborhood computation where points in each non-overlapping voxel in a regular grid are assigned to the same class by considering features within a support region defined by the voxel itself. The contribution provides offline training and online classification procedures as well as five alternative feature vector definitions based on principal component analysis for scatter, tubular and planar shapes. Moreover, the feasibility of this approach is evaluated by implementing a neural network (NN) method previously proposed by the authors as well as three other supervised learning classifiers found in scene processing methods: support vector machines (SVM), Gaussian processes (GP), and Gaussian mixture models (GMM). A comparative performance analysis is presented using real point clouds from both natural and urban environments and two different 3D rangefinders (a tilting Hokuyo UTM-30LX and a Riegl). Classification performance metrics and processing time measurements confirm the benefits of the NN classifier and the feasibility of voxel-based neighborhood.

show abstract

Role and task allocation framework for Multirobot collaboration with latent knowledge estimation

Gianni

Uddin

2020

Engineering Reports

Self Cite

View full text Add to dashboard Cite

In this work a novel framework for modeling role and task allocation in Cooperative Heterogeneous Multi-Robot Systems (CHMRSs) is presented. This framework encodes a CHMRS as a set of multidimensional relational structures (MDRSs). This set of structure defines collaborative tasks through both temporal and spatial relations between processes of heterogeneous robots. These relations are enriched with tensors which allow for geometrical reasoning about collaborative tasks. A learning schema is also proposed in order to derive the components of each MDRS. According to this schema, the components are learnt from data reporting the situated history of the processes executed by the team of robots. Data are organized as a multirobot collaboration treebank (MRCT) in order to support learning. Moreover, a generative approach, based on a probabilistic model, is combined together with nonnegative tensor decomposition (NTD) for both building the tensors and estimating latent knowledge. Preliminary evaluation of the performance of this framework is performed in simulation with three heterogeneous robots, namely, two Unmanned Ground Vehicles (UGVs) and one Unmanned Aerial Vehicle (UAV).

show abstract

Real-time autonomous 3D navigation for tracked vehicles in rescue environments

Cited by 26 publications

References 18 publications

3D localization, mapping and path planning for search and rescue operations

3D localization, mapping and path planning for search and rescue operations

Voxel-Based Neighborhood for Spatial Shape Pattern Classification of Lidar Point Clouds with Supervised Learning

Role and task allocation framework for Multirobot collaboration with latent knowledge estimation

Contact Info

Product

Resources

About