Coal and Gangue Recognition Method Based on Local Texture Classification Network for Robot Picking

Coal gangue is a kind of industrial waste in the coal mine preparation process. Compared to conventional manual or machine-based separation technology, vision-based methods and robotic grasping are superior in cost and maintenance. However, the existing methods may have a poor recognition accuracy problem in diverse environments since coals and gangues’ apparent features can be unreliable. This paper analyzes the current methods and proposes a vision-based coal and gangue recognition model LTC-Net for separation systems. The preprocessed full-scale images are divided into n × n local texture images since coals and gangues differ more on a smaller scale, enabling the model to overcome the influence of characteristics that tend to change with the environment. A VGG16-based model is trained to classify the local texture images through a voting classifier. Prediction is given by a threshold. Experiments based on multi-environment datasets show higher accuracy and stability of our method compared to existing methods. The effect of n and t is also discussed.

Target Localization and Power Allocation Using Wireless Energy Harvesting Sensors

As radio-frequency (RF) based wireless energy harvesting technology can provide remote and continuous power to low-power devices, e.g., wireless sensors, it may be a substitute for batteries and extend the lifetime of the wireless sensor networks. In this paper, we propose a wireless energy harvesting localization system (WEHLoc), which contains batteryless wireless sensors as anchors and an energy access point (E-AP) to transfer power to the anchors. We consider a passive target localization scenario, in which the anchors monitor the target and send the sensed ranging data back to the E-AP. Additionally, we formulate the optimal estimation accuracy problem which is a 0–1 mixed-integer programming problem and relates to the energy beam, target transmitted power, and deployed anchor density. Then, we develop the power allocation scheme of the E-AP to solve the objective. In order to reduce the complexity, we propose a heuristic method that converts the maximum estimation accuracy problem into the energy efficiency problem and use linear programming to solve them. The simulations demonstrate that WEHLoc can be massively deployed in a wide area, and the estimation error and the power consumption are relatively low.

A gravitational torque-compensated 2-DOF planar robotic arm design and its active control

Serial robot manipulators have their servo motors with reduction gears on the link joints. When it comes to hyper-redundant robots, this kind of joint actuation mechanism cannot be implemented since this makes hyper-redundant robots too heavy. Instead, cable driven mechanisms are preferred. However, the positioning accuracy is negatively affected by the cables. This paper addresses the positioning accuracy problem of cable driven hyper-redundant robots by employing a 2-DOF robotic arm whose modules are counter-balanced. While the actuators connected to the base actively do most of the work using cables and springs, light and compact actuators connected to the links produce precise motion. The method will result in compact, light and precise hyper-redundant robotic arms. The above-mentioned procedure governed by a control software including a 2D simulator developed is experimentally proved to be a feasible method to compensate the gravitational torque successfully.

DWM: A Decomposable Winograd Method for Convolution Acceleration

Winograd's minimal filtering algorithm has been widely used in Convolutional Neural Networks (CNNs) to reduce the number of multiplications for faster processing. However, it is only effective on convolutions with kernel size as 3x3 and stride as 1, because it suffers from significantly increased FLOPs and numerical accuracy problem for kernel size larger than 3x3 and fails on convolution with stride larger than 1. In this paper, we propose a novel Decomposable Winograd Method (DWM), which breaks through the limitation of original Winograd's minimal filtering algorithm to a wide and general convolutions. DWM decomposes kernels with large size or large stride to several small kernels with stride as 1 for further applying Winograd method, so that DWM can reduce the number of multiplications while keeping the numerical accuracy. It enables the fast exploring of larger kernel size and larger stride value in CNNs for high performance and accuracy and even the potential for new CNNs. Comparing against the original Winograd, the proposed DWM is able to support all kinds of convolutions with a speedup of ∼2, without affecting the numerical accuracy.

Sparse Representation for Different Animal Vertebra Classification along the Fixation Trajectory of Pedicle Screw

Pedicle screw (PS) implantation is an ideal method for the treatment of severe multilevel vertebral instability. The key problem is the accuracy of PS fixation. In this paper, the spectrum of different tissues along the fixation trajectory of PS is studied to tackle the accuracy problem. Fresh porcine vertebrae, bovine vertebrae, and ovine vertebrae were measured by using the near-infrared spectrum (NIRs) device to obtain the reflected spectrum from these vertebrae. Along the fixation trajectory of PS, the classification method based on the sparse representation-based classifier (SRC) was applied to different vertebral tissues (cortical bones and cancellous bones). Considering the large amount of spectral data, sparse preserving projection (SPP) was applied to improve the performance of SRC. The proposed method based on the SPP method for dimensionality reduction and the SRC method for tissue recognition was first used in vertebrae classification and showed superior performance compared with other classification methods, such as SVM and 1NN. The results gained from this project are vital significant to the development of hi-tech medical instruments with independent intellectual property rights.

Linear Pinch Equilibrium of Non-Neutral Plasma Revisited: Phenomenological Consequences of a Numerical Accuracy Problem

Weibel in 1959 under considerations of a collisionless non-neutral cylindrical plasma column studied a linear pinch confinement equilibrium. As reported here, due to non-linearity of the ordinary differential equations obtained for the electrostatic and magnetostatic fields is possible to demonstrate that the confining features previously obtained are extremely dependent on the initial conditions, and the arrangement of two parameters (β - the ratio between ion and electron mass; M/KT - ratio between relativistic rest energy associated with the pair electron-ion and thermal energy kT ) related to the plasma column characteristics. We investigated in this paper the plasma column behavior (confining or non-confining) under modifications of that set of parameters. We detected a set of parameters values that imposes a confining configuration with an electronic skin effect on the plasma column, not yet reported or discussed in the literature.<br /><br />