Irregular-shaped Objects Research Articles

Cardboard packaging minimization: Autonomous box selection by RGB-depth vision system and complementary deep learning classification

Amidst the dramatic expansion of online shopping, wasted cardboard packaging has become a significant challenge in modern-day society. In this paper, we propose an innovative approach to address the issue of box packaging waste by efficiently selecting an optimal cardboard box for a set of items. Employing depth vision and deep learning classification techniques, the proposed system analyzes depth images to compute an optimal cuboid for unpackaged items while accounting for fragility characteristics. The system, facilitated by two Intel RealSense RGB-D cameras, first captures the overhead and sideview images to determine the dimensions of an unknown object. Then, the developed vision system utilizes averaging and morphological opening filters to denoise the acquired images, calculates the difference between the background image and object image, and employs thresholding by Otsu’s algorithm to find the minimum cuboid of the irregular object. After, a two-step deep learning model is used to determine the general category and characteristics of an object, considering factors, such as fragility, that may necessitate wrapping or separate packaging. Finally, a modified largest area first fit algorithm uses the outputted dimensions for irregular-shaped objects and packaged objects to determine the optimal fitting box from the standard Amazon box catalog. The computer simulation shows that the proposed method can be implemented in a cardboard packaging warehouse for optimal box selection, effectively minimizing cardboard waste.

UrbanSegNet: An urban meshes semantic segmentation network using diffusion perceptron and vertex spatial attention

Urban meshes semantic segmentation is essential for comprehending the 3D real-world environments, as it plays a vital role across various application domains, including digital twins, 3D navigation, and smart cities. Nevertheless, the inherent topological complexities of urban meshes impede the precise representation of dependencies and local structures, yielding compromised segmentation accuracy, especially for small or irregularly-shaped objects like vegetation and vehicles. To address this challenge, we introduce UrbanSegNet, a novel end-to-end model incorporating diffusion perceptron blocks and a vertex spatial attention mechanism. The diffusion perceptron blocks can dynamically enlarge receptive fields to capture features from local to completely global, enabling effective representation of urban meshes using multi-scale features and increasing small and irregularly-shaped object segmentation accuracy. The vertex spatial attention mechanism extracts the internal correlations within urban meshes to enhance semantic segmentation performance. Besides, a tailored loss function is designed to enhance overall performance further. Comprehensive experiments on two datasets demonstrate that the proposed method outperforms the state-of-the-art models in terms of mean F1 score, recall, and mean intersection over union (mIoU). The experimental results also demonstrate that UrbanSegNet achieves higher segmentation accuracy on vehicles and high vegetation compared to the state-of-the-art methods, highlighting the superiority of our proposed model in extracting features of small and irregularly-shaped objects.

Heuristics Integrated Deep Reinforcement Learning for Online 3D Bin Packing

Online 3D Bin Packing Problem (3D-BPP) has a wide range of industrial applications and there is an emerging research interest in learning optimal bin packing policy and deploying it for real logistics applications. From the heuristic methods to the deep reinforcement learning (DRL) methods, the previous works have proposed many solutions to solve the online 3D-BPP. However, none of them have studied what and how heuristics can be modelled into DRL to build a more effective and practical bin packing pipeline. In this work, we thoroughly investigate what heuristics can be used in online 3D-BPP and how to effectively integrate the heuristics with the DRL. First, we design 3 different heuristics based on the physical rules of the real world and the experiences of the human packers, including the Physics-Heuristics, the Packing-Heuristics and the Unpacking-Heuristics. Second, we model the 3 types of heuristics into the DRL framework and propose a novel heuristic DRL method to solve the online 3D-BPP. Extensive experimental results show that our method achieves state-of-the-art bin packing performance and the resulting real-world system is able to reliably finish the bin packing task in real logistics scenarios. Supplementary video is available at https://www.youtube.com/watch?v=x8GpmEELq18. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The rapid growth of e-commerce has significantly increased the burden of human packers in logistic warehouses, where the workers need to pick the products from a conveyor and pack them into bins (i.e. the online 3D bin packing). Thus it is of great importance to develop intelligent robotic systems to replace human labor, which is a long-standing topic in the field of control and automation science. This paper makes a substantial contribution to the related field by studying the online 3D bin packing in terms of both the theory and practice. On the one hand, the simulated experiments suggest that the presented algorithm significantly improves the space utilization of bin packing. On the other hand, the robotic system developed based on the proposed method can favourably finish the bin packing task in real logistics scenarios, demonstrating the practical use of our approach. Consequently, the approach proposed in this paper is totally applicable in logistic warehouses and is promising to drastically improve the working efficiency of the product packing in real warehouses. In the future, we will extend the presented approach to pack irregular-shaped objects and then facilitate more logistics applications.

Convergence of the DDA for ensembles of objects of irregular shape

The discrete dipole approximation (DDA) is commonly used to compute light-scattering properties of irregularly shaped particles. The DDA maps the particle into an array of cubic cells with side d < λ. For a randomly oriented irregularly shaped particle, DDA has been shown accurate when kd|m| ≤ 1, where k is the wavenumber and m is the particle refractive index. We demonstrate that the DDA yields robust results even when kd|m| ≈ 1.2 when applied to ensembles of irregularly shaped dielectric particles and kd|m| ≈ 1.3 for conductive particles. This finding can greatly reduce the computational load required for performing such computations.

Жадная эвристика размещения ортогональных многогранников для оптимизированного решения задач компоновки объектов нерегулярной формы

The article deals with the cutting and packing problems of irregular shape objects, which consist in finding the most compact way to place a given set of objects of arbitrary geometry inside a certain limited space. These problems belong to the class of ­NP-hard discrete optimization problems for which there are no methods of polynomial complexity to obtain exact solutions, so in practice they are most often solved approximately using heuristic and metaheuristic optimization methods. When placing objects of irregular shape, it is additionally necessary to take into account the geometry of objects to determine the correctness of their placement relative to each other. Existing methods of analyzing the geometry of objects and the formed placement scheme, based on the use of phi-functions and the construction of a hodograph of a vector function of dense placement, theoretically provide the possibility of obtaining an accurate solution, but require the use of time-consuming methods of nonlinear optimization. Therefore, in order to increase the speed of packing a large number of irregular-shaped objects, their shape is transformed by voxelization, followed by combining the resulting set of voxels into orthogonal polyhedra. To improve the quality of the solutions obtained, the paper proposes a greedy heuristic for the placement of orthogonal polyhedra, which implements the choice of the best orientation option for the object being placed, in which the formed packing will be the densest in comparison with other available orientation options for this object. The analysis of the effectiveness of this greedy heuristic on the problems of irregular cutting and packing of three-dimensional objects is carried out. Computational experiments have shown that the proposed greedy heuristic provides very fast high-quality solutions. Additionally, the results of testing the greedy placement heuristics using a genetic algorithm to optimize solutions to the packing problem are presented.

Object measurement in real underwater environments using improved stereo matching with semantic segmentation

Accurately measuring distances in underwater environments is critical for applications such as underwater exploration and maintenance of underwater structures, but is limited by underwater environmental factors. To address this challenge, semantic segmentation and visual stereo matching are promising methods to achieve this goal. In this paper, we proposed a combination of Retinex and Curvature filter for image enhancement, and a modified SGBM (Semi-Global Block Matching) method using dilated SAD(sum of absolute differences) window and the dynamic disparity range obtained using semantic segmentation results to accelerate stereo matching, while also providing insight into the content of the sensed scene. By using the method we proposed, the measurement error is significantly reduced, even in extreme environments with low visibility and complex backgrounds. It showed strong adaptability when dealing with irregular shape objects, and sped up the calculation speed of the algorithm by reducing the calculation area. In the comparison experiment, we found that the error of SGBM with our image enhancement was reduced to 22.8%. By using our stereo matching algorithm, the error was further reduced to 6.2%, demonstrating its superiority for underwater measurement tasks. Our approach offered an efficient solution for autonomous distance measurement in challenging underwater environments.

Movement Monitoring System for a Pneumatic Muscle Actuator

Recent advancements in soft pneumatic robot research have demonstrated these robots’ capability to interact with the environment and humans in various ways. Their ability to move over rough terrain and grasp objects of irregular shape, regardless of position, has garnered significant interest in developing new pneumatic soft robots. Integrating industrial design with related technologies holds great promise for the future, potentially bringing about a new lifestyle and revolutionizing the industry. As robots become increasingly practical, there is a growing need for sensitivity, robustness, and efficiency improvements. It is anticipated that the development of these intelligent pneumatic soft robots will play a critical role in serving the needs of society and production shortly. The present article is concerned with developing a system for monitoring a pneumatic robot’s parameters, including a spatial coordinate system. The focus is on utilizing the relationship between the coordinates and pressure to model the movement of the soft robot within the MATLAB simulation environment.

High-resolution model of complexly shaped bodies motion using an IBM-VOF-DEM coupling method

A new coupling model that focuses on the simulation of gas-liquid flow and irregular-shaped particulate flows was proposed based on the coupling of FD/IBM, VOF and DEM. A wide range of computing cases was presented for validation, including the spherical and cylindrical particle settling in fluid, where the particle Reynold's number is small, and the vortex shedding phenomenon could still be captured. Further, the simulations that the wooden blocks float upward and the cones fall into the water at a high Reynolds number were constructed. All these simulating results keep well consistent with the experiments. It turns out that the model is reliable for simulating the particulate flows as Reynold's number is up to 420,000. Besides, the model shows great strength of the solid-fluid motions capture capability, especially for the irregular-shaped objects moving trajectories at different scales from 0.00167 to 0.3 m, which is potentially helpful in industrial engineering. • A fully rsolved IB-VOF-DEM model is developed. • Typical and multi-scaling validating simulations are presented. • The model covers a wide range of Reynolds number from 400 to 420,000. • Multi-sphere model and Monte-Carlo method are employed.

Stimuli‐Responsive Sponge for Imaging and Measuring Weak Compression Stresses

Imaging and measuring compression stresses secure a safe and healthy life. Compression stresses in kPa range are not easily detected by conventional mechanoresponsive materials because microscopic molecular motion of the chromophores is not induced by such weak stresses. Moreover, imaging of the stress distribution is not achieved so far. The present study shows a sponge device combining two stimuli-responsive materials, a capsule releasing interior liquid and color-changing polymer in responses to compression stress and chemical stimulus, respectively. The stimuli-responsive capsule is dispersed on a melamine sponge comprised of the fibers with coating the layered polydiacetylene (PDA). The application of weak compression stresses induces collapse of the capsules, outflow of the interior liquid, and subsequent irreversible color change of PDA. The cascading response in the sponge device colorimetrically enables imaging of the distribution and measuring the strength of the compression stresses in kPa range. Furthermore, the device demonstrates imaging and measuring unknown weak compression stresses applied by the irregular-shaped objects. A couple of clinical issues in surgical operation of intestine are studied using the stress-imaging sponge device. The device and its design strategy can be applied to stress imaging in a variety of fields.

Convex-Hull Feature Adaptation for Oriented and Densely Packed Object Detection

Detecting oriented and densely packed objects is a challenging problem considering that the receptive field intersection between objects causes spatial feature aliasing. In this paper, we propose a convex-hull feature adaptation (CFA) approach, with the aim to configure convolutional features in accordance with irregular object layouts. CFA roots in the convex-hull feature representation, which defines a set of dynamically sampled feature points guided by the convex intersection over union (CIoU) to bound object extent. CFA pursues optimal feature assignment by constructing convex-hull sets and iteratively splitting positive or negative convex-hulls. By simultaneously considering overlapping convex-hulls and objects and penalizing convex-hulls shared by multiple objects, CFA defines a systematic way to adapt convolutional features on regular grids to objects of irregular shapes. Experiments on DOTA and SKU110K-R datasets show that CFA achieved new state-of-the-art performance for detecting oriented and densely packed objects. CFA also sets a solid baseline for convex polygon prediction on the MS COCO dataset defined for general object detection. Code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/SDL-GuoZonghao/BeyondBoundingBox</uri> .

Design a Multifunctional Soft Tactile Sensor Enhanced by Machine Learning Approaches

Abstract Tactile sensors are essential to robot hands that deal with various objects and interact with the environment. Soft tactile sensors are especially important to capture tactile information about delicate, irregular-shaped, or unknown objects. This paper introduces a soft tactile sensor that can simultaneously estimate the contact force, contact feature, and contact point. Inspired by multifunctional human skin, the proposed design has a dual-layer structure and is multifunctional. The top layer consists of a group of sensing elements that detect the contact location and contact feature enhanced by the bagging classifier based on the k-nearest neighbors. The sensor elements were biomimetically and analytically designed as a pyramid shape that mimicked the mountain ridge-like structure in human skin to improve sensitivity. The bottom layer was made by a piece of Velostat sandwiched between conductive fabrics that can measure the contact force. The relationship between the sensing voltage and the contact force was modeled by the Nadaraya–Watson regressor. The performance of the proposed sensor was verified by a repeatability test. Furthermore, we demonstrated the effectiveness of the proposed sensor on a robotic gripper. The experimental results show that this sensor is able to detect contact information of fragile objects.

Acoustic Beam Splitting and Cloaking Based on a Compressibility-Near-Zero Medium

We report an artificial acoustic compressibility-near-zero medium made of a phononic crystal composed of epoxy blocks arranged in a square lattice. Its anisotropic effective density leads to a linear cross in its isofrequency contour in the vicinity of the Brillouin zone center, as its effective compressibility approaches zero. When a Gaussian beam is normally incident on the phononic crystal, a splitting effect is achieved at the frequency of the crossing point. Based on such a beam-splitting effect, an acoustic cloaking of an irregular-shaped object embedded in the phononic crystal is demonstrated both theoretically and experimentally. Such an anisotropic zero-index material offers a potential method to control acoustic waves.

Topology Optimization Design and Experiment of a Soft Pneumatic Bending Actuator for Grasping Applications

Soft pneumatic bending actuators are commonly used in robotic grasping applications and can be applied for handling both delicate and irregularlyshaped objects. Because these soft pneumatic actuators are typically embedded with multiple air chambers inside soft, thin structures, their maximum payloads are usually low compared to rigid designs of similar size. This article presents a soft pneumatic bending actuator design using a topology optimization procedure that can maximize the bending capability and thus increase the allowable payload of soft grippers that consist of multiple pneumatic bending actuators. After an optimum design is obtained, multiple identical actuators are prototyped through a molding process using a silicone rubber material. Both two-fingered and three-fingered soft grippers are developed using the topology-optimized pneumatic bending actuators. Experiments, including a bending angle test, an output force test, and a payload test, are conducted in this study,and the results are compared against the test results of the commercial SRT soft pneumatic actuator. The experimental results show that the bending angle for the developed bending actuator is 111 degrees on average of 10 tests at an input pressure of 50 kPa. The output force is 9.45 N at an input pressure of 80 kPa, while the maximum payloads of the two-fingered and three-fingered grippers are 2.66 kg and 5.12 kg, respectively.

Development of algorithms for the formation and placement of N-dimensional orthogonal polyhedrons into containers of complex geometric shape

The article is devoted to problems related with the development of algorithms for creating, optimizing and placing of orthogonal polyhedrons of arbitrary dimension. Under N-dimensional orthogonal polyhedron is understood a composite object represented as a set of N-dimensional parallelepipeds with a fixed position relative to each other, which is considered as a whole. The need to solve the problem of placing a set of objects in the form of orthogonal polyhedrons of arbitrary dimension arises when solving a large number of practical problems in the automation of design and production. In particular, solving the layout problems of equipment and space, problems of cutting materials, problems of designing very large-scale integrated circuits, and optimizing the distribution of resources in computing and production systems, as well as many other actual problems, can be reduced to solving the problem of packing orthogonal polyhedrons. To create a new N-dimensional orthogonal polyhedron, the operations of addition, subtraction and multiplication of orthogonal objects are implemented. With the aim to increase the placement speed of orthogonal polyhedrons which were created with using the voxel model, an algorithm for partitioning of an orthogonal polyhedron into many non-overlapping each other as large as possible large orthogonal objects is proposed. In the scientific literature, there are no solutions to the problem of partitioning an orthogonal polyhedron of arbitrary dimension. The proposed partitioning algorithm is based on the usage of a model of potential containers developed to solve the orthogonal packing problems of arbitrary dimension. To obtain a container of arbitrary shape, a possibility of setting its geometric constraints with an orthogonal polyhedron is implemented. The article presents an algorithm developed for packing an N-dimensional orthogonal polyhedrons inside containers of arbitrary geometric shape, based on the application of multiplication orthogonal polyhedrons operations to obtain the set of points of permissible placement of each considered object. The examples of solving the problems of cutting the industrial materials into objects of irregular shapes, as well as solving the problems of placement complex objects created with using the voxel model into containers of various geometric shapes (including sphere and cone) are given. The effectiveness of the application of the developed algorithms is shown on the example of solving the problem of generation a dense layout of details on a 3D printer platform.

Automatic Generation Method for Training Dataset of Instance Segmentation for Irregularly Shaped Object Picking

In the field of robotics, robotic bin picking has been studied extensively. In this paper, we consider an irregularly shaped object as a grasping target. For detecting irregularly shaped objects, instance segmentation based on deep learning is one of the promising methods. One of the challenges for deep learning-based methods is how to reduce the time and effort to prepare the dataset for training. In this paper, we propose a method to automatically generate a dataset for learning instance segmentation using only information available from public image databases. The proposed method achieves mean average precision (mAP) of 0.85 for the automatically generated test data. It also showed mAP of 0.65 for the test data generated using untrained irregularly shaped objects, and achieved a success rate of more than 98% in picking experiments with the robot.

Advanced Progress Control of Infrastructure Construction Projects Using Terrestrial Laser Scanning Technology

Effective progress control is vital for steering infrastructure construction to completion with minimum delay. Walking through the infrastructure project site to record progress in different activities is time-consuming, requiring information extracted from construction drawings, schedules, and budgets, as well as data collected from the construction site. This process can be automated by using advanced remote sensing technologies. This study contributes to progress monitoring in large horizontal infrastructure projects. It presents a practical automated method using laser scanning technology that can track the project’s progress in a real construction environment with limited human input. It is robust and accurate and is currently operational. The system capitalizes on the success of laboratory experiments. This system deals with occlusions effectively, accelerates the registration process of multiple scans, reduces the noise in the data, recognizes the objects of irregular shape, and is economically feasible. It provides evidence that all current challenges encountered in using laser scanners in monitoring construction progress can be overcome. This method has been successfully tested in the Wacker Drive reconstruction project in Chicago, IL.

Application of the Nesvetay Code for Solving Three-Dimensional High-Altitude Aerodynamics Problems

A survey of the capabilities of the Nesvetay code as applied to computing the flow of a high-speed monatomic gas around objects of irregular shape for large flight altitudes is given. An implicit numerical method on an arbitrary unstructured grid and a two-level approach to the organization of parallel computations are described. This code is compared with the well-known MONACO and SMILE codes that implement the direct simulation Monte Carlo method.

BIM reconstruction from 3D point clouds: A semantic registration approach based on multimodal optimization and architectural design knowledge

Reconstructing semantically rich building information model (BIM) from 2D images or 3D point clouds represents a research realm that is gaining increasing popularity in architecture, engineering, and construction. Researchers have found that architectural design knowledge, such as symmetry, planarity, parallelism, and orthogonality, can be utilized to improve the effectiveness of such BIM reconstruction. Following this line of enquiry, this paper aims to develop a novel semantic registration approach for complicated scenes with repetitive, irregular-shaped objects. The approach first formulates the architectural repetition as the multimodality in mathematics. Thus, the reconstruction of repetitive objects becomes a multimodal optimization (MMO) problem of registering BIM components which have accurate geometries and rich semantics. Then, the topological information about repetition and symmetry in the reconstructed BIM is recognized and regularized for BIM semantic enrichment. A university lecture hall case, consisting of 1.9 million noisy points of 293 chairs, was selected for an experiment to validate the proposed approach. Experimental results showed that a BIM was satisfactorily created (achieving about 90% precision and recall) automatically in 926.6 s; and an even more satisfactory BIM achieved 99.3% precision and 98.0% recall with detected semantic and topological information under the minimal effort of human intervention in 228.4 s. The multimodality model of repetitive objects, the repetition detection and regularization for BIM, and satisfactory reconstruction results in the presented approach can contribute to methodologies and practices in multiple disciplines related to BIM and smart city.

Large Sample Neutron Activation Analysis of a Peruvian Pottery Artifact

The present work was carried out within the first Large Sample Neutron Activation Analysis (LSNAA) intercomparison study organized under the auspices of the International Atomic Energy Agency (IAEA). Replicates of a Peruvian pottery object, representing a Pre-Columbian archaeological artifact, were prepared by the Instituto Peruano de Energia Nuclear (IPEN) and distributed to 10 laboratories with LSNAA capa- bilities. The LSNAA results were compared against values derived by conventional Instrumental Neutron Activation Analysis (INAA) on small amounts of the material used to produce the test object. The results of the intercomparison study will be presented in detail in a separate publication. In this work, the LSNAA methodology developed at ’Demokritos’ for analysis of large and irregular shaped objects is discussed. An accurate description of the complex geometry of the object was obtained by Computerized Tomography X-ray scanning. The correction factors required for the effects of neutron self-shielding and gamma-ray self-attenuation within the large sample material were derived by Monte Carlo simulations using MCNP code. The results of the study showed that LSNAA provides elemental concentration values with a satisfactory agreement against values obtained by conventional INAA. Therefore, LSNAA is a technique suitable for the purpose of analysis of intact pottery objects of irregular shape for archaeological studies.

Total surface area in indoor environments.

Certain processes in indoor air, such as deposition, partitioning, and heterogeneous reactions, involve interactions with surfaces. We have characterized the surface area, volume, shape, and material of objects in 10 bedrooms, nine kitchens, and three offices. The resolution of the measurements was ∼1 cm. The ratio of surface area with contents to that without contents did not vary by type of room and averaged 1.5 ± 0.3 (mean ± standard deviation) across all rooms. The ratio of the volume minus contents to nominal volume averaged 0.9 ± 0.1 and was lower for kitchens compared to bedrooms and offices. Ignoring contents, the surface-area-to-volume ratio was 1.8 ± 0.3 m-1; accounting for contents, the ratio was 3.2 ± 1.2 m-1, or 78% higher. These two ratios did not vary by type of room and were similar to those measured for 33 rooms in another study. Due to substantial differences in the design and contents of kitchens, their ratios had the highest variability among the three room types. The most common shape of surfaces was flat rectangular, while each room also had many irregularly-shaped objects. Paint-covered surfaces and stained wood were the two most common materials in each room, accounting for an average of 42% and 22% of total surface area, respectively, although the distribution of materials varied by room type. These findings have important implications for understanding the chemistry of indoor environments, as the available surface area for deposition, partitioning, and reactions is higher and more complex than assumed in simple models.