The risk of getting psychiatric disease or various kinds of dementia is rising as a significant problem in the aging society. Functional near-infrared spectroscopy (fNIRS) can measure the blood chromophores noninvasiv...
详细信息
ISBN:
(纸本)9781728130583
The risk of getting psychiatric disease or various kinds of dementia is rising as a significant problem in the aging society. Functional near-infrared spectroscopy (fNIRS) can measure the blood chromophores noninvasively for early diagnosis, and frequent examination, which are vital in case of brain degeneration. We present the development and functioning of our lab-developed fNIRS system. It provides a real-time display of variation in blood chromophores related to neuronal activation. It employs 128 dual-wavelength LEDs of 735 nm and 850 nm. The selection of these wavelengths allows computation of oxyhemoglobin (HbO) and deoxyhemoglobin (HbR). A single photosensor is used in the presented device. This system is developed using a modular approach where a single module can cover approximately 7 cm x 7 cm while multiple modules can be used to cover a wider area. The current configuration utilizes different source-detector separation to reach multiple depths between 2 cm and 3.5 cm. Short separation channels also exist in the design to provide the information of superficial layers. MOSFET based LED switching is implemented that allows sharp current switching for high-speed data acquisition. Windows-based software is developed for the display of fNIRS data in real time. Wi-Fi is used as the wireless medium of communication between the hardware and software. Phantom model, as well as human subject, was used for testing the device efficacy. The phantom results showed that by increasing the channel-separation, the signal intensity was reduced. Resting state human subject was also evaluated to compute and display the HbO in real time. A complete fNIRS sample comprising of 128 channels was recorded in 25 ms.
Soft robots are inherently safe and comply readily to their environment. They are therefore exciting for applications like search and rescue or medicine, which involve a high degree of uncertainty, and require interac...
ISBN:
(数字)9781728165707
ISBN:
(纸本)9781728165714
Soft robots are inherently safe and comply readily to their environment. They are therefore exciting for applications like search and rescue or medicine, which involve a high degree of uncertainty, and require interacting with humans. However, the best way to model and control soft robots largely remains an open question. One promising approach is to leverage physically-based modeling techniques such as the finite element method. However, such techniques are inherently limited by their physical assumptions. Indeed, real-world soft robots are often made from unpredictable materials, using imprecise techniques. Data-driven approaches provide an exciting alternative, as they can learn real-world fabrication defects and asymmetries. In this paper we present our first investigation into using machine learning to do soft robot control. We learn a differentiable model of a soft robot's quasi-static physics, and then perform gradient-based optimization to find optimal open-loop control inputs. We find that our learned model captures phenomena that would be absent from an idealized physically-based simulation. We also present practical techniques for acquiring high-quality motion capture data, and observations the effect of network complexity on model accuracy.
In this paper, we present a Cloud robotics (CR) platform for performing environmental monitoring in data centers. Due to the high energy consumption, the need of monitoring for controlling air temperature in each data...
详细信息
Due to the nonlinearity and complexity of uncertain multi-finger robot, the ordinary host system is difficult to control effectively. An uncertain multi-finger robot host system based on multiprocessor (MP) and LM alg...
详细信息
In this paper, an adhesion system used for magnetic wall climbing robots using permanent magnets is examined for its change in adhesion concerning the inter-magnet distance within the Yoke. Pocket milled ferrous mater...
详细信息
ISBN:
(数字)9781728170893
ISBN:
(纸本)9781728170909
In this paper, an adhesion system used for magnetic wall climbing robots using permanent magnets is examined for its change in adhesion concerning the inter-magnet distance within the Yoke. Pocket milled ferrous material of high permeability named yoke is used to hold neodymium magnets in mobile wall-climbing robots. For a climbing robot, the self-weight is a significant parameter which needs to be at the minimum possible value. This simulation study is expected to help in bringing down the weight of the yoke holding magnets and help in finding the right inter magnet distance for the magnets used. Simulation studies were carried out for various inter magnet distances and the values are reported. An optimized distance was arrived using the simulation results thus obtained.
Video action recognition is widely applied in video indexing, intelligent surveil-lance, multimedia understanding, and other fields. Recently, it was greatly improved by incorporating the learning of deep information ...
详细信息
ISBN:
(纸本)9781450365307
Video action recognition is widely applied in video indexing, intelligent surveil-lance, multimedia understanding, and other fields. Recently, it was greatly improved by incorporating the learning of deep information using convolutional neural network (ConvNet). In this paper, we proposed a 3D ConvNet-GRU architecture to learn deep information for action recognition. Specifically, we use 3D ConvNet to learn spatiotemporal information from short RGB clips and optical flow clips, and impose gated recurrent unit (GRU) on the spatiotemporal information to model the temporal evolution for action recognition. The experimental results show that our 3D ConvNet-GRU method is effective to model temporal evolution for action and achieves recognition performance comparable to that of state-of-the-art methods.
Inter-camera pedestrians association always employs appearance features to merge tractlets of the same pedestrian into a whole. However, appearance features are always view-and illumination-sensitive. In this paper, w...
详细信息
ISBN:
(纸本)9781450365307
Inter-camera pedestrians association always employs appearance features to merge tractlets of the same pedestrian into a whole. However, appearance features are always view-and illumination-sensitive. In this paper, we present a method to solve inter-camera pedestrian association via discriminative learned feature in a stable way with illumination transfer. First, we proposed a discriminative feature learning model which is a convolution siamese network that combines the verification and identification losses. Furthermore, we introduce color brightness transfer reduce color distortions under different illumination. To learn proper brightness transfer function, a fuzzy color cluster is used to model the change of color brightness between different cameras. The experiments show the effectiveness of the proposed method and achieve the state-of-the-art in the benchmark NLRP_MCT dataset.
The 3D virtual scenes are widely used in different applications, such as visual simulation, augmented reality and virtual reality and so on. For reducing the time consumption with 3D modeling, we propose an improved m...
详细信息
ISBN:
(纸本)9781538695944
The 3D virtual scenes are widely used in different applications, such as visual simulation, augmented reality and virtual reality and so on. For reducing the time consumption with 3D modeling, we propose an improved method of 3D scene reconstruction based on SfM. By taking the video streaming as input, we put forward a feature similarity determination strategy to extract keyframes, and utilize a dense algorithm to improve the model accuracy. Moreover, the method appends 3D model filtering to remove the redundancy of the result models. We demonstrate our method by testing the time cost and effect of different videos and datasets. Results illustrate that our method is practical and available.
The increasing labor shortage issue and the working safety awareness cause the urgent of the development of new construction methods. A new type of lab for new construction processes is required to expedite the innova...
详细信息
ISBN:
(纸本)9780784482438
The increasing labor shortage issue and the working safety awareness cause the urgent of the development of new construction methods. A new type of lab for new construction processes is required to expedite the innovation cycle. This paper presents an ongoing work of building a construction lab at University of Alberta. The goal of the construction lab is to provide a sandbox for developing new construction processes and machines. We designed the lab with four major systems: (1) sensors: to collect data from construction site for operation assistants and virtual reconstruction;(2) manipulators: to excavate the path planning algorithms and to develop the cooperation approaches between human and machines;(3) visualizers: to construct the digital twin of a real construction site for revealing the simulated results in virtual environment;(4) computers: to run machine learning algorithms for recognizing and tracking objects in construction environments. This laboratory allows the researchers in the construction engineering test and develop their tools in a controlled environment. Such scaled tests in the lab can bring significant benefits in finance, efficiency, and safety. Technical Area: robotics, automation, and control. Application Context: Project design, construction, planning, and management.
Soft strain sensors are becoming increasingly popular for obtaining tactile information in soft robotic applications. Diverse technological solutions are being investigated to design these sensors. Simultaneously, new...
ISBN:
(数字)9781728165707
ISBN:
(纸本)9781728165714
Soft strain sensors are becoming increasingly popular for obtaining tactile information in soft robotic applications. Diverse technological solutions are being investigated to design these sensors. Simultaneously, new methods for modeling these sensor are being proposed due to their highly nonlinear, time varying properties. Among them, machine learning based approaches, particularly using dynamic recurrent neural networks look the most promising. However, these complex networks have large number of free parameters to be tuned, making it difficult to apply them for real-world applications. This paper introduces the concept of transfer learning for modelling soft strain sensors, which allows us to utilize information learned in one task to be applied to another task. We demonstrate this technique on a passive anthropomorphic finger with embedded strain sensors used for two regression tasks. We show how the transfer learning approach can drastically reduce the number of free parameters to be tuned for learning new skills. This work is an important step towards scaling of sensor networks (algorithm-wise) and for using soft sensor data for high-level control tasks.
暂无评论