Position In 3D Space Research Articles

Point-to-Spike Residual Learning for Energy-Efficient 3D Point Cloud Classification

Spiking neural networks (SNNs) have revolutionized neural learning and are making remarkable strides in image analysis and robot control tasks with ultra-low power consumption advantages. Inspired by this success, we investigate the application of spiking neural networks to 3D point cloud processing. We present a point-to-spike residual learning network for point cloud classification, which operates on points with binary spikes rather than floating-point numbers. Specifically, we first design a spatial-aware kernel point spiking neuron to relate spiking generation to point position in 3D space. On this basis, we then design a 3D spiking residual block for effective feature learning based on spike sequences. By stacking the 3D spiking residual blocks, we build the point-to-spike residual classification network, which achieves low computation cost and low accuracy loss on two benchmark datasets, ModelNet40 and ScanObjectNN. Moreover, the classifier strikes a good balance between classification accuracy and biological characteristics, allowing us to explore the deployment of 3D processing to neuromorphic chips for developing energy-efficient 3D robotic perception systems.

Visual guidance fine-tunes probing movements of an insect appendage

Visually guided reaching, a regular feature of human life, comprises an intricate neural control task. It includes identifying the target's position in 3D space, passing the representation to the motor system that controls the respective appendages, and adjusting ongoing movements using visual and proprioceptive feedback. Given the complexity of the neural control task, invertebrates, with their numerically constrained central nervous systems, are often considered incapable of this level of visuomotor guidance. Here, we provide mechanistic insights into visual appendage guidance in insects by studying the probing movements of the hummingbird hawkmoth's proboscis as they search for a flower's nectary. We show that visually guided proboscis movements fine-tune the coarse control provided by body movements in flight. By impairing the animals' view of their proboscis, we demonstrate that continuous visual feedback is required and actively sought out to guide this appendage. In doing so, we establish an insect model for the study of neural strategies underlying eye-appendage control in a simple nervous system.

Research on Visualization Teaching of Art and Design in Colleges and Universities Combining 3D Modeling Technology

Abstract This paper discusses constructing 3D modeling process for art and design works using 3D display technology and SIFT feature algorithm. Firstly, the pictures with different angles are obtained by changing the geometric information of the graphics and determining their positions in 3D space. Then the lighting model and texture mapping are used to enhance the realism, and the feature points are extracted in combination with the SIFT algorithm to determine the angle of the pictures. An initial 3D model is generated by image silhouette convergence, which corrects the image segmentation several times to complete the modeling. In addition, the study applies the modeling technique to the visualization teaching of art design, and verifies the teaching effect through students’ attitude and practical performance. The results show that most students have positive attitudes towards visualization teaching, especially in Class 2 and Class 15, where the favorite rate reaches 65%. Under this teaching method, students can flexibly utilize the design elements to create diversified art works.

Data Driven 3D-Lane Detection Using Parallelism Loss Function

Accurate lane position prediction is crucial in autonomous driving for safe vehicle maneuvering. Monocular cameras, aided by AI advancements, have proven to be effective in this task. However, 2D image space predictions overlook lane height, causing poor results in uphill or downhill scenarios that affect action judgments, such as in the planning and control module. Previous 3D-lane detection approaches relied solely on applying Inverse Perspective Mapping (IPM) on the encoded camera feature map, which may not be ordered according to the perspective principle leading to sub-optimal prediction results. To address these issues, we present the LS-3DLane network, inspired by the Lift-Splat-Shoot architecture, which predicts lane position in 3D space using a data-driven approach. The network also employs the Parallelism loss, using prior knowledge of lane geometry, to improve performance. Such loss can be used for training any 3D lane position prediction network and would boost the performance. Our results show that LS-3DLane outperforms previous approaches like Gen-LaneNet and 3D-LaneNet, with F-score improvements reaching 5.5% and 10%, respectively, in certain cases. LS-3DLane performs similarly in X/Z error metrics. Parallelism loss was shown to boost the F-Score KPI when applied to any of the models under test (LS-3DLane, GenLaneNet, and 3D-LaneNet) by up to 2.8% in certain cases and has a positive impact on nearly all the other KPIs.

Accurate classification of major brain cell types using in vivo imaging and neural network processing.

Comprehensive analysis of tissue cell type composition using microscopic techniques has primarily been confined to ex vivo approaches. Here, we introduce NuCLear (Nucleus-instructed tissue composition using deep learning), an approach combining in vivo two-photon imaging of histone 2B-eGFP-labeled cell nuclei with subsequent deep learning-based identification of cell types from structural features of the respective cell nuclei. Using NuCLear, we were able to classify almost all cells per imaging volume in the secondary motor cortex of the mouse brain (0.25 mm3 containing approximately 25,000 cells) and to identify their position in 3D space in a noninvasive manner using only a single label throughout multiple imaging sessions. Twelve weeks after baseline, cell numbers did not change yet astrocytic nuclei significantly decreased in size. NuCLear opens a window to study changes in relative density and location of different cell types in the brains of individual mice over extended time periods, enabling comprehensive studies of changes in cell type composition in physiological and pathophysiological conditions.

A Novel Analytical Model for Mutual Inductance Calculations Between Two Nonidentical N-Sided Polygonal Planar Coils Arbitrarily Positioned in 3-D Space for Wireless Power Transfer

Mutual inductance (MI) calculation is of great importance, because there are many systems that operate on the principle of inductive coupling. This paper presents a novel analytical model that calculates the MI between two non-identical n-sided polygonal planar coils located in arbitrary position in 3D space. The model depends upon misalignment of both the primary and the secondary coils of a wireless power transfer (WPT) system operating based on inductive coupling. The polygonal coils used in this model may differ from each other in terms of size, shape and turns number. The misalignment cases considered in the proposed model include lateral, angular, lateral-angular and vertical distance variation between the primary and secondary coils. Based on Stokes Theorem, instead of using the magnetic flux density vector, the magnetic vector potential (MVP) is used to reach the required magnetic flux in the calculation of the MI. A single analytical solution function that can calculate the MI between two planar polygonal coils of any shape in a simple manner and short time duration is obtained by performing difficult mathematical operations through MVP. Results obtained from the developed new model are compared with finite element analysis (FEA), experimental and other studies in the literature. The good agreement of the results of this study with the other results has proved the accuracy and validity of the proposed new analytical model.

Flying enhances viewing from above bias on ambiguous visual stimuli.

The human spatial orientation system is well designed on the ground but is imperfect in the aeronautical three-dimensional (3D) environment. However, human perception systems perform Bayesian statistics based on encountered environments and form shortcuts to improve perceptual efficiency. It is unknown whether our perception of spatial orientation is modified by flying experience and forms perceptual biases. The current study tested pilot perceptual biases on ambiguous visual stimuli, the bistable point-light walkers, and found that flying experiences increased the pilot's tendency to perceive himself as higher than the target and the target as farther away from them. Such perceptual effects due to flight are likely to be attributed to experience of variable vestibular state in a higher position in 3D space, rather than the experience of a higher viewpoint. Our findings suggest that flying experience will modifies our visual perceptual biases, and that more attention should be paid to the enhanced viewing from above bias when flying to avoid overestimating altitude or viewing angle when the visual conditions are ambiguous.

Embedded Magnetic Sensing for Feedback Control of Soft HASEL Actuators

The need to create more viable soft sensors is increasing in tandem with the growing interest in soft robots. Several sensing methods, like capacitive stretch sensing and intrinsic capacitive self-sensing, have proven to be useful when controlling soft electro-hydraulic actuators, but are still problematic. This is due to challenges around high-voltage electronic interference or the inability to accurately sense the actuator at higher actuation frequencies. These issues are compounded when trying to sense and control the movement of a multiactuator system. To address these shortcomings, we describe a two-part magnetic sensing mechanism to measure the changes in displacement of an electro-hydraulic (HASEL) actuator. Our magnetic sensing mechanism can achieve high accuracy and precision for the HASEL actuator displacement range, and accurately tracks motion at actuation frequencies up to 30 Hz, while being robust to changes in ambient temperature and relative humidity. The high accuracy of the magnetic sensing mechanism is also further emphasized in the gripper demonstration. Using this sensing mechanism, we can detect submillimeter difference in the diameters of three tomatoes. Finally, we successfully perform closed-loop control of one folded HASEL actuator using the sensor, which is then scaled into a deformable tilting platform of six units (one HASEL actuator and one sensor) that control a desired end effector position in 3D space. This work demonstrates the first instance of sensing electro-hydraulic deformation using a magnetic sensing mechanism. The ability to more accurately and precisely sense and control HASEL actuators and similar soft actuators is necessary to improve the abilities of soft, robotic platforms.

Manipulating propagation and evolution of polarization singularities in composite Bessel-like fields

Structured optical fields embedded with polarization singularities (PSs) have attracted extensive attention due to their capability to retain topological invariance during propagation. Many advances in PS research have been made over the past 20 years in the areas of mathematical description, generation and detection technologies, propagation dynamics, and applications. However, one of the most crucial and difficult tasks continues to be manipulating PSs with multiple degrees of freedom, especially in three-dimensional (3D) tailored optical fields. We propose and demonstrate the longitudinal PS lines obtained by superimposing Bessel-like modes with orthogonal polarization states on composite vector optical fields (VOFs). The embedded PSs in the fields can be manipulated to propagate robustly along arbitrary trajectories, or to annihilate, revive, and transform each other at on-demand positions in 3D space, allowing complex PS’ topological morphology and intensity patterns to be flexibly customized. Our findings could spur further research into singular optics and help with applications such as micromanipulation, microstructure fabrication, and optical encryption.

Mechanistic study of Hg(II) interaction with three different α-aminophosphonate adsorbents: Insights from batch experiments and theoretical calculations.

Herein, efficient and potential chelating α-aminophosphonate based sorbents (AP-) derived from three different amine origins (aniline/anthranilic acid/O-phenylenediamine) to form AP-H, carboxylated and aminated enhanced aminophosphonate as AP-H, AP-COOH, and AP-NH2 were synthesized via a facile method. The structure of the synthesized sorbents was elucidated using different techniques; elemental analysis (CHNP/O), FT-IR, NMR (1H-, 13C and 31P NMR), TGA and BET. The fabricated sorbents were exploited for Hg(II) removal from aqueous solution via sorption properties. Isotherm fitted by Langmuir equation: the maximum sorption capacities at optimum pH 5.5, and T:25±1°C, were found to be 1.33, 1.23, and 1.15mmol Hg g-1 for AP-COOH, AP-NH2, AP-H, respectively, which is roughly correlated with the active sites density and the hard/soft characteristics of adsorbents' reactive groups. Metal-ligand binding affinities are qualitatively rationalized in terms of hard and soft acids and bases (HSAB) theory. The interaction of Hg(II) (soft) has a stronger affinity to AP-COOH can be considered a softer base compared with reference material (AP-H) over than AP-NH2 (hard). This sequence result showed opposite trends consistent with their reciprocal properties according to the steric effect modulates and the specific surface area. Thermodynamics analysis for absolute values of ΔH°, ΔS° and ΔG° afford the selectivity towards Hg(II) sorption with the following order: AP-COOH>AP-NH2 >AP-H. Elution and regeneration was carried out by HCl solution and recycled for a minimum of five cycles, the sorption and desorption efficiencies are greater than 91%. Such sorbents exhibit good durability, stability and promising potential for Hg(II) removal. Finally, a new modelling technique for quantitative non-linear description and comparison of equivalent geographical positions in 3D space of extended relationships. Exothermic and spontaneous behavior were observed using a proposed Floatotherm that included the Van't Hoff parameters model.

ToyAssembler: Sewing Patterns into a 3D Toy by Automatic Pre-Positioning

In this paper, we present an efficient tool ToyAssembler for sewing 2D patterns into 3D plush/inflatable toys. To simulate a 3D toy from 2D patterns, it is necessary to arrange 2D patterns at appropriate positions in 3D space according to their sewing relations. However, manual pre-positioning is time-consuming and hard to achieve satisfactory results. ToyAssembler provides an automatic pattern pre-positioning algorithm. Firstly, a sewing relation graph is constructed based on sewing correspondences, then patterns are divided into several groups by decomposing the graph. Secondly, patterns of each group are wrapped onto a cylinder, and then all groups of wrapped patterns are assembled to obtain the final 3D patterns with appropriate positions for sewing. At last, pre-positioned patterns are sewn into a 3D toy using physically-based simulation. Experiment results demonstrate that patterns are reasonably pre-positioned and satisfactory toy models can be created. Our approach significantly reduces the time and effort of rebuilding toy models from 2D patterns. • A novel method is presented for sewing 2D patterns into 3D plush/inflatable toy models by automatic pre-positioning. • A pattern grouping method is proposed by decomposing a sewing relation graph. • A wrapping and assembling method is proposed to place patterns at appropriate positions for sewing simulation.

Benefits of active listening during 3D sound localization

In everyday life, sound localization entails more than just the extraction and processing of auditory cues. When determining sound position in three dimensions, the brain also considers the available visual information (e.g., visual cues to sound position) and resolves perceptual ambiguities through active listening behavior (e.g., spontaneous head movements while listening). Here, we examined to what extent spontaneous head movements improve sound localization in 3D—azimuth, elevation, and depth—by comparing static vs. active listening postures. To this aim, we developed a novel approach to sound localization based on sounds delivered in the environment, brought into alignment thanks to a VR system. Our system proved effective for the delivery of sounds at predetermined and repeatable positions in 3D space, without imposing a physically constrained posture, and with minimal training. In addition, it allowed measuring participant behavior (hand, head and eye position) in real time. We report that active listening improved 3D sound localization, primarily by ameliorating accuracy and variability of responses in azimuth and elevation. The more participants made spontaneous head movements, the better was their 3D sound localization performance. Thus, we provide proof of concept of a novel approach to the study of spatial hearing, with potentials for clinical and industrial applications.

Designation of the Quality of EGNOS+SDCM Satellite Positioning in the Approach to Landing Procedure

The main aim of this paper is to present the results of research on the application of a modified mathematical model to determine the quality parameters of SBAS (Satellite Based Augmentation System) satellite positioning in aviation. The authors developed a new calculation strategy to determine the resultant values of the parameters of accuracy, continuity, availability and integrity of SBAS positioning. To achieve it, a weighted mean model was used for the purposes of developing a mathematical algorithm to determine the resultant values of SBAS positioning. The created algorithm was implemented for two SBAS supporting systems, i.e., EGNOS (European Geostationary Navigation Overlay Service) and SDCM (System of Differential Correction and Monitoring). The algorithm was tested in a flight test conducted with a Diamond DA 20-C airplane in north-eastern Poland in 2020. The conducted research revealed that the resultant error of the position in 3D space determined with use of the proposed weighted mean model improved by, respectively, 1–7% in comparison to the standard arithmetic mean model and by 1–37% in comparison to a single SBAS/EGNOS solution. Moreover, the application of the Multi-SBAS positioning algorithm results in an increase in the nominal results of continuity and availability by 50% in comparison to the arithmetic mean model. Apart from that, the values of the integrity parameters determined with use of the proposed weighted mean model improved by 62–63% in comparison to the standard arithmetic mean model.

Multiarray Eigenbeam-ESPRIT for 3D Sound Source Localization With Multiple Spherical Microphone Arrays

A 3D sound source localization (SSL) technique for multiple spherical arrays is proposed. Spherical arrays have been popularly used for direction of arrival (DoA) estimation, which serves as important prior knowledge for source separation, dereverberation, and binaural rendering for VR applications. In this work, a parametric 3D SSL technique is proposed that can detect multiple sources using multiple spherical array recordings. From the enriched information by using multiple arrays, we show that the parametric estimation of 3D positions is possible without scanning candidate positions in 3D space. To mitigate the problem of the ambiguous association between DoAs estimated from different arrays, a total covariance matrix including both auto- and cross-covariance matrices between arrays is utilized such that DoAs from arrays are automatically paired to uniquely localize 3D source positions. The joint parameter estimation is based on the generalized joint Schur decomposition of multiple matrices combined with a geometric projection robustly guiding the convergence of the proposed iteration algorithm. Simulations conducted in anechoic and reverberant room conditions reveal that the proposed technique can accurately determine multiple source positions in 3D space without ambiguities.

An efficient three-dimensional indoor visible light positioning algorithm using multiple transmitters and single receiver

Indoor visible light positioning system has received much attention in recent years due to the simultaneous operation of optical wireless communication and object tracking. Here, a 3D positioning algorithm is proposed using multiple transmitters and single receiver. The algorithm is based on dividing the 3D space in several 2D planes along the room height maintaining small resolution and further compute the received signal acts as a reference power for each of the 3D coordinates. By keeping such reference power, the algorithm continues the operation to search for receiver’s unknown 3D position based on the best match of power levels within the set tolerance value. The investigation results show that the average positioning error is 0.658 cm at 3 m height considering a standard room dimension. Moreover, the proposed algorithm is also used for trajectory tracking of a random curve, where around 92% of the positioning error is maintained below 2.7 cm. By knowing the geometry of a room with static setup of transmitters, the proposed algorithm is easy to implement and converge within a short time to get the receiver’s unknown position in 3D space.

RUBubbles as a novel tool to study categorization learning

Grouping objects into discrete categories affects how we perceive the world and represents a crucial element of cognition. Categorization is a widespread phenomenon that has been thoroughly studied. However, investigating categorization learning poses several requirements on the stimulus set in order to control which stimulus feature is used and to prevent mere stimulus–response associations or rote learning. Previous studies have used a wide variety of both naturalistic and artificial categories, the latter having several advantages such as better control and more direct manipulation of stimulus features. We developed a novel stimulus type to study categorization learning, which allows a high degree of customization at low computational costs and can thus be used to generate large stimulus sets very quickly. ‘RUBubbles’ are designed as visual artificial category stimuli that consist of an arbitrary number of colored spheres arranged in 3D space. They are generated using custom MATLAB code in which several stimulus parameters can be adjusted and controlled separately, such as number of spheres, position in 3D-space, sphere size, and color. Various algorithms for RUBubble generation can be combined with distinct behavioral training protocols to investigate different characteristics and strategies of categorization learning, such as prototype- vs. exemplar-based learning, different abstraction levels, or the categorization of a sensory continuum and category exceptions. All necessary MATLAB code is freely available as open-source code and can be customized or expanded depending on individual needs. RUBubble stimuli can be controlled purely programmatically or via a graphical user interface without MATLAB license or programming experience.Graphical abstract

Marsbot: A monopod robot capable of achieving three-dimensional dynamic and stable jumping

Continuous, stable and accurate jumping in three-dimensional (3D) environment is one of difficulties of jumping robot research. In this paper, a monopod jumping robot Marsbot whose center of mass (CoM) can accurately reach desired positions in 3D space is designed. Based on the spring loaded inverted pendulum (SLIP) dynamics model, the take-off velocity control of Marsbot was realized, which made controllable continuous jump possible. Based on the reaction wheel pendulum (RWP) dynamics model and the law of conservation of angular momentum, the real-time 3D attitude control of Marsbot with three inertial tails was realized. By integrating SLIP model, RWP model and the post-landing steering strategy proposed in this paper, a continuous jump algorithm for CoM of Marsbot to accurately reach desired 3D positions is proposed. This paper proposes the air target grasping strategy of Marsbot (the author will elaborate on this issue in another article): When the robot jumps to the desired grasping position in the air, the robotic arm can quickly and stably grasp the target object such as tree branches, so that the robot can perch or hang on the target object, and get a good view from high place. The simulation results show that Marsbot can achieve continuous, stable, accurate jumping and realize air perching/observation operations through the above schemes. The simulation results also verify feasibility of the 3D jump dynamics model and its control algorithm proposed in this paper.

Implementation of a System for Real-Time Detection and Localization of Terrain Objects on Harvested Forest Land

Research highlights: An automatic localization system for ground obstacles on harvested forest land based on existing mature hardware and software architecture has been successfully implemented. In the tested area, 98% of objects were successfully detected and could on average be positioned within 0.33 m from their true position in the full range 1–10 m from the camera sensor. Background and objectives: Forestry operations in forest environments are full of challenges; detection and localization of objects in complex forest terrains often require a lot of patience and energy from operators. Successful automatic real-time detection and localization of terrain objects not only can reduce the difficulty for operators but are essential for the automation of harvesting and logging tasks. We intend to implement a system prototype that can automatically locate ground obstacles on harvested forest land based on accessible hardware and common software infrastructure. Materials and Methods: An automatic object detection and localization system based on stereo camera sensing is described and evaluated in this paper. This demonstrated system detects and locates objects of interest automatically utilizing the YOLO (You Only Look Once) object detection algorithm and derivation of object positions in 3D space. System performance is evaluated by comparing the automatic detection results of the tests to manual labeling and positioning results. Results: Results show high reliability of the system for automatic detection and location of stumps and large stones and shows good potential for practical application. Overall, object detection on test tracks was 98% successful, and positional location errors were on average 0.33 m in the full range from 1–10 m from the camera sensor. Conclusions: The results indicate that object detection and localization can be used for better operator assessment of surroundings, as well as input to control machines and equipment for object avoidance or targeting.

Spatial point pattern methodology for the study of pores 3D patterning in two casting aluminium alloys

Two cast aluminium alloys fabricated by different casting processes (gravity die-casting and lost foam casting) and showing different degrees of porosity were characterized with X-ray Computed tomography. Information concerning the pore distribution inside the investigated materials is obtained in terms of pore size and the pore positions in 3D space. Subsequently, a spatial point pattern analysis is undertaken to investigate the pore distributions. Different methods, including a nearest neighbor analysis, Ripley's K-function and Clark-Evans tests developed for 3D applications, are used to analyse the observed patterns. The results show that the Homogeneous Poisson process, which provides the Complete Spatial Randomness (CSR) is suitable to approximate the spatial distribution of the pores present in the investigated alloys. Synthetic microstructures that mimic key macroscale features of the materials in terms of pore size and the 3D spatial distribution of the pores were generated. These microstructures can be used in the probabilistic modelling of fatigue behavior.

Stocktaking Robots, Automatic Inventory, and 3D Product Maps: The Smart Warehouse Enabled by UHF-RFID Synthetic Aperture Localization Techniques

In our age of global networks and the Internet of Things, communication and identification with a transponder tagged on goods and products have become increasingly important [1]. There is not only a need for the identification of the labeled objects but also for finding their exact positions in 3D space. Radio-frequency identification (RFID) offers both and is, in fact, already used in a vast field of applications, such as access control, goods logistics, indoor localization, automation, stocktaking, and many others [2]-[6]. Figure 1 illustrates a mobile robot platform for automatic inventory rounds that is used to acquire 3D product maps with ultrahigh-frequency (UHF)-RFID technology.