The integration of computervisiontechniques for the accomplishment of autonomous interaction tasks represents a challenging research direction in the context of aerial robotics. In this paper, we consider the proble...
详细信息
ISBN:
(数字)9781728162126
ISBN:
(纸本)9781728162133
The integration of computervisiontechniques for the accomplishment of autonomous interaction tasks represents a challenging research direction in the context of aerial robotics. In this paper, we consider the problem of contact-based inspection of a textured target of unknown geometry and pose. Exploiting state of the art techniques in computer graphics, tuned and improved for the task at hand, we designed a framework for the projection of a desired trajectory for the robot end-effector on a generically-shaped surface to be inspected. Combining these results with previous work on energy-based interaction control, we are laying the basis of what we call vision-based impedance control paradigm. To demonstrate the feasibility and the effectiveness of our methodology, we present the results of both realistic ROS/Gazebo simulations and preliminary experiments with a fully-actuated hexarotor interacting with heterogeneous curved surfaces whose geometric description is not available a priori, provided that enough visual features on the target are naturally or artificially available to allow the integration of localization and mapping algorithms.
3D vision systems will play an important role in next-generation dairy farming due to the sensing capabilities they provide in the automation of animal husbandry tasks such as the monitoring, herding, feeding, milking...
详细信息
3D vision systems will play an important role in next-generation dairy farming due to the sensing capabilities they provide in the automation of animal husbandry tasks such as the monitoring, herding, feeding, milking and bedding of animals This paper will review 3D computervision systems and techniques that are and may be implemented in Precision Dairy Farming. This review will include evaluations of the applicability of Time of Flight and Streoscopic vision systems to agricultural applications as well as a breakdown of the categories of computervisionalgorithms which are being explored in a variety of use cases. These use cases range from robotic platforms such as milking robots and autonomous vehicles which must interact closely and safely with animals to intelligent systems which can identify dairy cattle and detect deviations in health indicators such as Body Condition Score and Locomotion Score. Upon analysis of each use case, it is apparent that systems which can operate in unconstrained environments and adapt to variations in herd characteristics, weather conditions, farmyard layout and different scenarios in animal-robot interaction are required. Considering this requirement, this paper proposes the application of techniques arising from the emerging field of research in Artificial Intelligence that is Geometric Deep Learning. (C) 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
Scene understanding represents one of the most primary problems in computervision. It implies the full knowledge of all the elements of the environment and the comprehension of the relationships between them. One of ...
详细信息
ISBN:
(纸本)9781728136059
Scene understanding represents one of the most primary problems in computervision. It implies the full knowledge of all the elements of the environment and the comprehension of the relationships between them. One of the major tasks in this process is the scene recognition, on which we focus in this work. Scene recognition is a relevant and helpful task in many robotic fields such as navigation, localization, manipulation, among others. The knowledge of the place (e.g. "office", "classroom" or "kitchen") can improve the performance of robots in indoor environments. This task can be difficult because of the variability, ambiguity, illumination changes, occlusions and scale variability present in this type of spaces. Commonly, this problem has been approached through the development of models based on local and global characteristics, incorporating context information and, more recently, using deep learning techniques. In this paper, we propose a multi-classifier model for scene recognition considering as priors the outcomes of independent base classifiers. We implement a weighted voting scheme based on genetic algorithms for the combination of different classifiers in order to improve the recognition performance. The results have proved the validity of our approach and how the proper combination of independent classifier models makes it possible to find a better and more efficient solution for the scene recognition problem.
Event cameras are an emerging technology in computervision, offering extremely low latency and bandwidth, as well as a high temporal resolution and dynamic range. Inherent data compression is achieved as pixel data i...
详细信息
ISBN:
(纸本)9781538680940
Event cameras are an emerging technology in computervision, offering extremely low latency and bandwidth, as well as a high temporal resolution and dynamic range. Inherent data compression is achieved as pixel data is only produced by contrast changes at the edges of moving objects. However, current trends in state-of-the-art visual algorithms rely on deep-learning with networks designed to process colour and intensity information contained in dense arrays, but are notoriously computationally heavy. While the combination of these visual technologies could lead to fast, efficient, and accurate detection and recognition algorithms, it is uncertain whether the compressed event-camera data actually contain the required information for these techniques to discriminate between objects and a cluttered background. This paper presents a pilot study in which off-the-shelf deep-learning is applied to visual events for object detection on the iCub robotic platform, and analyses the impact of temporal integration of the event data. We also present a novel pipeline that bootstraps event-based dataset annotation from mature frame-based algorithms, in order to more quickly generate the required datasets.
Absolute orientation estimation is the determination of the similarity transformation between two sets of corresponding 3D points, a task arising frequently in computervision and robotics. We have recently proposed a...
详细信息
ISBN:
(纸本)9781538680940
Absolute orientation estimation is the determination of the similarity transformation between two sets of corresponding 3D points, a task arising frequently in computervision and robotics. We have recently proposed an absolute orientation algorithm based on the Fast Optimal Attitude Matrix (FOAM) algorithm from astronautics and demonstrated that it is more efficient computationally compared to widely-used approaches involving costly eigen- and singular-value matrix decompositions. In this work, we compare our FOAM-based solution with several more algorithms derived from attitude estimation techniques and show that further computational savings are possible by employing an algorithm grounded on the Optimal Linear Attitude Estimator (OLAE) method.
The modern technological era is witnessing a significant advancement in AI related fields such as machine learning, mobile robots and autonomous vehicles. The success of such systems is immensely dependent on computer...
详细信息
ISBN:
(纸本)9781538644300
The modern technological era is witnessing a significant advancement in AI related fields such as machine learning, mobile robots and autonomous vehicles. The success of such systems is immensely dependent on computervisionalgorithms. The entirely software enabled intelligent vehicles will soon be hitting the roads in the coming decade. Such self-controlled mobile robots may still be vulnerable to accidents or crashes. Therefore, the industry requires some state of the art techniques to substantiate the safety and protection of its passengers as well as other road users. Simulation based testing methods have been in use from a long time, but the newer smart vehicle innovations require better versions of simulation techniques. We thus provide an improved mechanism for simulation testing to validate safe navigation of cars in variety of traffic scenarios. Neural network approach integrated with agent based modelling is described in this paper.
Teaching a robot to predict and mimic how a human moves or acts in the near future by observing a series of historical human movements is a crucial first step in human-robot interaction and collaboration. In this pape...
详细信息
ISBN:
(纸本)9781538680940
Teaching a robot to predict and mimic how a human moves or acts in the near future by observing a series of historical human movements is a crucial first step in human-robot interaction and collaboration. In this paper, we instrument a robot with such a prediction ability by leveraging recent deep learning and computervisiontechniques. First, our system takes images from the robot camera as input to produce the corresponding human skeleton based on real-time human pose estimation obtained with the OpenPose library. Then, conditioning on this historical sequence, the robot forecasts plausible motion through a motion predictor, generating a corresponding demonstration. Because of a lack of high-level fidelity validation, existing forecasting algorithms suffer from error accumulation and inaccurate prediction. Inspired by generative adversarial networks (GANs), we introduce a global discriminator that examines whether the predicted sequence is smooth and realistic. Our resulting motion GAN model achieves superior prediction performance to state-of-the-art approaches when evaluated on the standard H3.6M dataset. Based on this motion GAN model, the robot demonstrates its ability to replay the predicted motion in a human-like manner when interacting with a person.
The modern technological era is witnessing a significant advancement in AI related fields such as machine learning, mobile robots and autonomous vehicles. The success of such systems is immensely dependent on computer...
详细信息
The modern technological era is witnessing a significant advancement in AI related fields such as machine learning, mobile robots and autonomous vehicles. The success of such systems is immensely dependent on computervisionalgorithms. The entirely software enabled intelligent vehicles will soon be hitting the roads in the coming decade. Such self-controlled mobile robots may still be vulnerable to accidents or crashes. Therefore, the industry requires some state of the art techniques to substantiate the safety and protection of its passengers as well as other road users. Simulation based testing methods have been in use from a long time, but the newer smart vehicle innovations require better versions of simulation techniques. We thus provide an improved mechanism for simulation testing to validate safe navigation of cars in variety of traffic scenarios. Neural network approach integrated with agent based modelling is described in this paper.
In this paper, we design an autonomous flight controller for height regulation and bang-bang controller for directional control of a light-weight flapping-wing micro air vehicle (FWMAV) with limited payload. We also p...
详细信息
暂无评论