Human augmented mapping for indoor environments using a stereo camera

Kim, Soohwan; Cheong, Howon; Park, Ju-Hong; Park, Sung-Kee

doi:10.1109/iros.2009.5354128

Cited by 12 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same principle wherein explicit human feedback is given to the robot is also present within several other works [35,3,7,18,30,16], however, their strong dependence on human supervision constrains their operation alongside trained users and simplified scenarios.…”

Section: Introductionmentioning

confidence: 89%

Binding human spatial interactions with mapping for enhanced mobility in dynamic environments

Papadakis

Rives

2016

Auton Robot

View full text Add to dashboard Cite

To cite this version:Panagiotis Papadakis, Patrick Rives. Binding human spatial interactions with mapping for enhanced mobility in dynamic environments. Autonomous Robots, Springer Verlag, 2017, 41 (5) Abstract For mobile robots to operate in compliance with human presence, interpreting the impact of human activities and responding constructively is a challenging goal. In this paper, we propose a generative approach for enhancing robot mapping and mobility in the presence of humans through a joint, probabilistic treatment of static and dynamic characteristics of indoor environments. Human spatial activity is explicitly exploited for the purpose of passage detection and space occupancy prediction while effectively discarding false positive human detections using prior map information. In turn, this allows the execution of plan trajectories within unexplored areas by using human presence for resolving the uncertainty or ambiguity that is due to dynamic events. A series of experiments with an indoor robot navigating in close human proximity within a multifloor building demonstrate the effectiveness of our approach in realistic conditions.

show abstract

Section: Introductionmentioning

confidence: 89%

Binding human spatial interactions with mapping for enhanced mobility in dynamic environments

Papadakis

Rives

2016

Auton Robot

View full text Add to dashboard Cite

show abstract

“…Another important consideration for autonomous inspection of large components is the ability to operate in indoor scenarios with dynamic obstacles, such as humans. In [11], a quadrocopter equipped with a gimbal-mounted camera was used to perform autonomous inspection in indoor environments with obstacles, with the algorithm using a combination of visual odometry and collision avoidance algorithms to plan a dynamic flight path and avoid collisions. Furthermore, in [12], a quadrocopter equipped with a gimbal-mounted camera and a LIDAR sensor were used to perform autonomous inspection in indoor environments, with the sensor providing 3D mapping of the environment and helping to avoid collisions.…”

Section: Related Workmentioning

confidence: 99%

UAV Inspection of Large Components: Determination of Alternative Inspection Points and Online Route Optimization

Schörner

Wanninger

Katschinsky

et al. 2023

2023 IEEE/ACM 5th International Workshop on Robotics Software Engineering (RoSE)

View full text Add to dashboard Cite

Automation is playing an increasing role in the field of quality assurance. For the visual inspection of larger assemblies such as aircraft fuselages or ship hulls, the use of UAVs is an option. This paper deals with one aspect of the UAVsupported inspection of assemblies in production. Here, newly added components have to be checked for correct assembly. The planning of the shortest possible route from which all components to be inspected can be viewed as well as the estimation of the UAV's position relative to the component have already been presented in previous work. We propose strategies that can be used if an inspection point cannot be reached by the UAV or the component to be inspected cannot be seen by the UAV's camera from the inspection point. For this purpose, we generate alternative inspection points that can be used if errors occur during the inspection from the original inspection point. To achieve this, we present a metric that can be used to select an alternative inspection point that is as suitable as possible. We conclude by demonstrating how this strategy works by evoking different failure cases in a simulated environment.

show abstract

“…In contrast, most of the effort in R-HAM has been concentrated on either incorporating human input remotely via tele-operation such as in the Urban Search and Rescue (USAR) problem (Murphy 2004;Nourbakhsh et al 2005), or in high level decision making such as goal assignment or coordination of multiple agents (Olson et al 2013;Parasuraman et al 2007;Doroodgar et al 2010). Some R-HAM techniques for metric mapping and pose estimation have also been explored, but these involve either having the robot retrace its steps to fill in parts missed by the human (Kim et al 2009) or by having additional agents and sensors in the environment (Kleiner, Dornhege, and Dali 2007).…”

Section: Related Workmentioning

confidence: 99%

Human-in-the-Loop SLAM

Nashed

Biswas

2018

AAAI

View full text Add to dashboard Cite

Building large-scale, globally consistent maps is a challenging problem, made more difficult in environments with limited access, sparse features, or when using data collected by novice users. For such scenarios, where state-of-the-art mapping algorithms produce globally inconsistent maps, we introduce a systematic approach to incorporating sparse human corrections, which we term Human-in-the-Loop Simultaneous Localization and Mapping (HitL-SLAM). Given an initial factor graph for pose graph SLAM, HitL-SLAM accepts approximate, potentially erroneous, and rank-deficient human input, infers the intended correction via expectation maximization (EM), back-propagates the extracted corrections over the pose graph, and finally jointly optimizes the factor graph including the human inputs as human correction factor terms, to yield globally consistent large-scale maps. We thus contribute an EM formulation for inferring potentially rank-deficient human corrections to mapping, and human correction factor extensions to the factor graphs for pose graph SLAM that result in a principled approach to joint optimization of the pose graph while simultaneously accounting for multiple forms of human correction. We present empirical results showing the effectiveness of HitL-SLAM at generating globally accurate and consistent maps even when given poor initial estimates of the map.

show abstract

Human augmented mapping for indoor environments using a stereo camera

Cited by 12 publications

References 9 publications

Binding human spatial interactions with mapping for enhanced mobility in dynamic environments

Binding human spatial interactions with mapping for enhanced mobility in dynamic environments

UAV Inspection of Large Components: Determination of Alternative Inspection Points and Online Route Optimization

Human-in-the-Loop SLAM

Contact Info

Product

Resources

About