Search results

Traditional visual simultaneous localization and mapping (SLAM) systems are primarily based on the assumption that the environment is static, which makes them struggle with the interference caused by dynamic objects in complex industrial production environments. This paper aims to improve the stability of visual SLAM in complex dynamic environments through semantic segmentation and its optimization.

This paper proposes a real-time visual SLAM system for complex dynamic environments based on YOLOv5s semantic segmentation, named YLS-SLAM. The system combines semantic segmentation results and the boundary semantic enhancement algorithm. By recognizing and completing the semantic masks of dynamic objects from coarse to fine, it effectively eliminates the interference of dynamic feature points on the pose estimation and enhances the retention and extraction of prominent features in the background, thereby achieving stable operation of the system in complex dynamic environments.

Experiments on the Technische Universität München and Bonn data sets show that, under monocular and Red, Green, Blue - Depth modes, the localization accuracy of YLS-SLAM is significantly better than existing advanced dynamic SLAM methods, effectively improving the robustness of visual SLAM. Additionally, the authors also conducted tests using a monocular camera in a real industrial production environment, successfully validating its effectiveness and application potential in complex dynamic environment.

This paper combines semantic segmentation algorithms with boundary semantic enhancement algorithms to effectively achieve precise removal of dynamic objects and their edges, while ensuring the system's real-time performance, offering significant application value.

Assembly sequence planning (ASP) plays a vital role in assembly process because it directly influences the feasibility, cost and time of the assembly process. The purpose of this study is to solve ASP problem more efficiently than current algorithms.

A novel assembly subsets prediction method based on precedence graph is proposed to solve the ASP problem. The proposed method adopts the idea of local to whole and integrates a simplified firework algorithm. First, assembly subsets are generated as initial fireworks. Then, each firework explodes to several sparks with higher-level assembly subsets and new fireworks are selected for next generation according to selection strategy. Finally, iterating the algorithm until complete and feasible solutions are generated.

The proposed method performs better in comparison with state-of-the-art algorithms because of the balance of exploration (fireworks) and exploitation (sparks). The size of initial fireworks population determines the diversity of the solution, so assembly subsets prediction method based on precedence graph (ASPM-PG) can explore the solution space. The size of sparks controls the exploitation ability of ASPM-PG; with more sparks, the direction of a specific firework can be adequately exploited.

The proposed method is with simple structure and high efficiency. It is anticipated that using the proposed method can effectively improve the efficiency of ASP and reduce computing cost for industrial applications.

The proposed method finds the optimal sequence in the construction process of assembly sequence rather than adjusting order of a complete assembly sequence in traditional methods. Moreover, a simplified firework algorithm with new operators is introduced. Two basic size parameters are also analyzed to explain the proposed method.

Dynamic movement primitives (DMPs) is a general robotic skill learning from demonstration method, but it is usually used for single robotic manipulation. For cloud-based robotic skill learning, the authors consider trajectories/skills changed by the environment, rebuild the DMPs model and propose a new DMPs-based skill learning framework removing the influence of the changing environment.

The authors proposed methods for two obstacle avoidance scenes: point obstacle and non-point obstacle. For the case with point obstacles, an accelerating term is added to the original DMPs function. The unknown parameters in this term are estimated by interactive identification and fitting step of the forcing function. Then a pure skill despising the influence of obstacles is achieved. Using identified parameters, the skill can be applied to new tasks with obstacles. For the non-point obstacle case, a space matching method is proposed by building a matching function from the universal space without obstacle to the space condensed by obstacles. Then the original trajectory will change along with transformation of the space to get a general trajectory for the new environment.

The proposed two methods are certified by two experiments, one of which is taken based on Omni joystick to record operator’s manipulation motions. Results show that the learned skills allow robots to execute tasks such as autonomous assembling in a new environment.

This is a new innovation for DMPs-based cloud robotic skill learning from multi-scene tasks and generalizing new skills following the changes of the environment.

Assembly action recognition plays an important role in assembly process monitoring and human-robot collaborative assembly. Previous works overlook the interaction relationship between hands and operated objects and lack the modeling of subtle hand motions, which leads to a decline in accuracy for fine-grained action recognition. This paper aims to model the hand-object interactions and hand movements to realize high-accuracy assembly action recognition.

In this paper, a novel multi-stream hand-object interaction network (MHOINet) is proposed for assembly action recognition. To learn the hand-object interaction relationship in assembly sequence, an interaction modeling network (IMN) comprising both geometric and visual modeling is exploited in the interaction stream. The former captures the spatial location relation of hand and interacted parts/tools according to their detected bounding boxes, and the latter focuses on mining the visual context of hand and object at pixel level through a position attention model. To model the hand movements, a temporal enhancement module (TEM) with multiple convolution kernels is developed in the hand stream, which captures the temporal dependences of hand sequences in short and long ranges. Finally, assembly action prediction is accomplished by merging the outputs of different streams through a weighted score-level fusion. A robotic arm component assembly dataset is created to evaluate the effectiveness of the proposed method.

The method can achieve the recognition accuracy of 97.31% and 95.32% for coarse and fine assembly actions, which outperforms other comparative methods. Experiments on human-robot collaboration prove that our method can be applied to industrial production.

The author proposes a novel framework for assembly action recognition, which simultaneously leverages the features of hands, objects and hand-object interactions. The TEM enhances the representation of dynamics of hands and facilitates the recognition of assembly actions with various time spans. The IMN learns the semantic information from hand-object interactions, which is significant for distinguishing fine assembly actions.

In confronting the challenge of climate change and progressing towards dual carbon goals, China is actively implementing low-carbon city pilot policy. This paper aims to focus on the potential impact of this policy on enterprise green governance, aiming to promote the reduction and balance of carbon emissions.

Based on the panel data of China's large-scale industrial enterprises from 2007 to 2013, this paper uses the Difference-in-differences (DID) method to study the impact and path mechanism of the implementation of low-carbon city pilot policy on enterprise green governance. Heterogeneity analysis is used to compare the effects of low-carbon city pilot policy in different regions, different enterprises and different industries.

The low-carbon pilot can indeed effectively enhance corporate green governance, a conclusion that still holds after a series of robustness tests. The low-carbon city pilot policy mainly enhances enterprise green governance through two paths: an industrial structure upgrade and enterprise energy consumption, and it improves green governance by reducing enterprise energy consumption through industrial structure upgrade. The impact of low-carbon city pilot policy on enterprise green governance shows significant differences across different regions, different enterprises and different industries.

This paper examines the impact of low-carbon city pilot policy on enterprise green governance. However, due to availability of data, there are still some limitations to be further tackled. The parallel trend test in this paper shows that the pilot policy has a significant positive effect on the green governance of enterprises. However, due to serious lack of data in some years, the authors only selected the enterprise data of a shorter period as our experimental data, which leads the results to still have certain deficiencies. For the verification of the impact mechanism, the conclusions obtained in this paper are relatively limited. Although all the mechanism tests are passed, the reliability of the results still needs to be further tested through future data samples. In addition, as the pilot policy of low-carbon cities is still in progress, the policy can be tracked and analysed in the future as more data are disclosed, and further research can be carried out through dimensional expansion.

Low-carbon city pilot policy plays an important role in inducing the green governance of enterprises. Therefore, policy makers can continue to strengthen the construction of low-carbon city pilots by refining pilot experience, building typical cases, actively promoting pilot policy experience, expanding pilot scope and enhancing the implementation efficiency of pilot policy nationwide, which will contribute to the optimization and upgrading of the regional industrial structure at the urban level and will provide experience and reference for the synergistic implementation plan of pollution reduction and carbon reduction.

The impact of the low-carbon city pilot policy on enterprise green governance not only exists in two separate paths of urban industrial upgrading and enterprise energy consumption but also exists in a chain transmission path from macro to micro. The authors find that the effect value of each influence path is different, and there is an obvious leading influence path for the role of enterprise green governance. Therefore, in the process of implementing a low-carbon city pilot policy, policies should be designed specifically for different mechanisms. Moreover, complementing and coordinating several paths should be advocated to give full play to the green governance effect of enterprises brought by different paths and to further expand the scope of industries and enterprises where policies play a role.

To the best of the authors’ knowledge, for the first time, this paper connects macro mechanisms with micro mechanisms, discovering a macro-to-micro transmission mechanism in the process of low-carbon city pilot policy affecting enterprise green governance. That is, the low-carbon city pilot policy can facilitate industrial structure upgrading, resulting in reduced enterprise energy consumption, ultimately enhancing enterprise green governance.

The purpose of this study is to investigate the effect of nano ZrC particles on the mechanical and electrochemical corrosion properties of FeCrAl alloys, providing a beneficial reference basis for the development of high-performance carbide reinforced FeCrAl alloys with good mechanical and corrosion properties in the future.

Nano ZrC reinforced FeCrAl alloys were prepared by mechanical alloying and spark plasma sintering. Phases composition, tensile fractography, corrosion morphology and chemical composition of nano ZrC reinforced FeCrAl alloys were analyzed by X-ray diffraction, scanning electron microscopy and energy dispersive X-ray spectroscopy, respectively. Microhardness and tensile properties of nano ZrC reinforced FeCrAl alloys were investigated by mechanical testing machine and Vickers hardness tester. Electrochemical corrosion properties of nano ZrC reinforced FeCrAl alloys were investigated by electrochemical workstation in 3.5 wt.% NaCl solution.

The results showed that addition of nano ZrC can effectively improve the mechanical and corrosion properties. However, excessive nano ZrC could decrease the mechanical properties and reduce the corrosion resistance. In all the FeCrAl alloys, FeCrAl–0.6 wt.% ZrC alloy exhibits the optimum mechanical properties with an ultimate tensile strength, elongation and hardness of 990.7 MPa, 24.1% and 335.8 HV1, respectively, and FeCrAl–0.2 wt.% ZrC alloy has a lower corrosion potential (−0.179 V) and corrosion current density (2.099 µA/cm2) and larger pitting potential (0.497 V) than other FeCrAl–ZrC alloys, showing a better corrosion resistance.

Adding proper nano ZrC particles can effectively improve the mechanical and corrosion properties, while the excessive nano ZrC is harmful to the mechanical and corrosion properties of FeCrAl alloys, which provides an instruction to develop high-performance FeCrAl cladding materials.

Previous studies have mostly discussed the impact of environmental policy on enterprise innovation, but the discussion on how turbulence in environmental policy may affect firms' green innovation has been insufficient. This paper explores the effect of environmental policy uncertainty on corporate green innovation in the turnover of environmental protection officials (EPOT) context.

The authors manually collected the data on the EPOT of 280 Chinese prefecture-level cities, and used the Poisson regression model to conduct empirical analyses based on the panel data of 1472 Chinese listed manufacturing firms from 2008 to 2017.

The results show that environmental policy uncertainty leads firms to reduce their green patent applications only for green invention patent applications. Such an effect is more pronounced in non-state-owned enterprises (non-SOEs). In addition, when the new directors of the Ecology and Environmental Bureau take office through promotions or are no more than 55 years old, the negative effect is more obvious, but there is no significant difference regardless of whether new directors have worked in environmental protection departments.

First, this paper supplements the research on the antecedents of corporate green innovation from the perspective of environmental policy uncertainty and extends the applications of real options theory. Second, this paper expands the research on the government–business relationship from the EPOT perspective.

In order to reduce logistics transportation costs and respond to low-carbon economy, the purpose of this paper is to study the more practical and common simultaneous pickup and delivery vehicle routing problem, which considers the carbon tax policy. A low-carbon simultaneous pickup and delivery vehicle routing problem model is constructed with the minimum total costs as the objective function.

This study develops a mathematical optimization model with the minimum total costs, including the carbon emissions costs as the objective function. An adaptive genetic hill-climbing algorithm is designed to solve the model.

First, the effectiveness of the algorithm is verified by numerical experiments. Second, the research results prove that carbon tax mechanism can effectively reduce carbon emissions within effective carbon tax interval. Finally, the research results also show that, under the carbon tax mechanism, the effect of vehicle speed on total costs will become more obvious with the increase of carbon tax.

This paper only considers the weight of the cargo, but it does not consider the volume of the cargo.

Few studies focus on environmental issues in the simultaneous pickup and delivery problem. Thus, this paper constructs a green path optimization model, combining the carbon tax mechanism for the problem. This paper further analyzes the impact of carbon tax value on total costs and carbon emission; at the same time, the effect of vehicle speed on total cost is also analyzed.

This paper aims to focus on the basic principle of WO₃ gas sensors to achieve high gas-sensing performance with good stability and repeatability. Metal oxide-based gas sensors are widely used for monitoring toxic gas leakages in the environment, industries and households. For better livelihood and a healthy environment, it is extremely helpful to have sensors with higher accuracy and improved sensing features.

In the present review, the authors focus on recent synthesis methods of WO₃-based gas sensors to enhance sensing features towards toxic gases.

This work has proved that the synthesis method led to provide different morphologies of nanostructured WO₃-based material in turn to improve gas sensing performance along with its sensing mechanism.

In this work, the authors reviewed challenges and possibilities associated with the nanostructured WO₃-based gas sensors to trace toxic gases such as ammonia, H₂S and NO₂ for future research.