A Novel Obstacle Traversal Method for Multiple Robotic Fish Based on Cross-Modal Variational Autoencoders and Imitation Learning.

Precision control of multiple robotic fish visual navigation in complex underwater environments has long been a challenging issue in the field of underwater robotics. To address this problem, this paper proposes a multi-robot fish obstacle traversal technique based on the combination of cross-modal variational autoencoder (CM-VAE) and imitation learning. Firstly, the overall framework of the robotic fish control system is introduced, where the first-person view of the robotic fish is encoded into a low-dimensional latent space using CM-VAE, and then different latent features in the space are mapped to the velocity commands of the robotic fish through imitation learning. Finally, to validate the effectiveness of the proposed method, experiments are conducted on linear, S-shaped, and circular gate frame trajectories with both single and multiple robotic fish. Analysis reveals that the visual navigation method proposed in this paper can stably traverse various types of gate frame trajectories. Compared to end-to-end learning and purely unsupervised image reconstruction, the proposed control strategy demonstrates superior performance, offering a new solution for the intelligent navigation of robotic fish in complex environments.

Trajectory Tracking and Obstacle Avoidance of Robotic Fish Based on Nonlinear Model Predictive Control.

The attainment of accurate motion control for robotic fish inside intricate underwater environments continues to be a substantial obstacle within the realm of underwater robotics. This paper presents a proposed algorithm for trajectory tracking and obstacle avoidance planning in robotic fish, utilizing nonlinear model predictive control (NMPC). This methodology facilitates the implementation of optimization-based control in real-time, utilizing the present state and environmental data to effectively regulate the movements of the robotic fish with a high degree of agility. To begin with, a dynamic model of the robotic fish, incorporating accelerations, is formulated inside the framework of the world coordinate system. The last step involves providing a detailed explanation of the NMPC algorithm and developing obstacle avoidance and objective functions for the fish in water. This will enable the design of an NMPC controller that incorporates control restrictions. In order to assess the efficacy of the proposed approach, a comparative analysis is conducted between the NMPC algorithm and the pure pursuit (PP) algorithm in terms of trajectory tracking. This comparison serves to affirm the accuracy of the NMPC algorithm in effectively tracking trajectories. Moreover, a comparative analysis between the NMPC algorithm and the dynamic window approach (DWA) method in the context of obstacle avoidance planning highlights the superior resilience of the NMPC algorithm in this domain. The proposed strategy, which utilizes NMPC, demonstrates a viable alternative for achieving precise trajectory tracking and efficient obstacle avoidance planning in the context of robotic fish motion control within intricate surroundings. This method exhibits considerable potential for practical implementation and future application.

Celebrating Diversity With Subtask Specialization in Shared Multiagent Reinforcement Learning.

Subtask decomposition offers a promising approach for achieving and comprehending complex cooperative behaviors in multiagent systems. Nonetheless, existing methods often depend on intricate high-level strategies, which can hinder interpretability and learning efficiency. To tackle these challenges, we propose a novel approach that specializes subtasks for subgroups by employing diverse observation representation encoders within information bottlenecks. Moreover, to enhance the efficiency of subtask specialization while promoting sophisticated cooperation, we introduce diversity in both optimization and neural network architectures. These advancements enable our method to achieve state-of-the-art performance and offer interpretable subtask factorization across various scenarios in Google Research Football (GRF).

A Path-Planning Method Based on Improved Soft Actor-Critic Algorithm for Mobile Robots.

The path planning problem has gained more attention due to the gradual popularization of mobile robots. The utilization of reinforcement learning techniques facilitates the ability of mobile robots to successfully navigate through an environment containing obstacles and effectively plan their path. This is achieved by the robots' interaction with the environment, even in situations when the environment is unfamiliar. Consequently, we provide a refined deep reinforcement learning algorithm that builds upon the soft actor-critic (SAC) algorithm, incorporating the concept of maximum entropy for the purpose of path planning. The objective of this strategy is to mitigate the constraints inherent in conventional reinforcement learning, enhance the efficacy of the learning process, and accommodate intricate situations. In the context of reinforcement learning, two significant issues arise: inadequate incentives and inefficient sample use during the training phase. To address these challenges, the hindsight experience replay (HER) mechanism has been presented as a potential solution. The HER mechanism aims to enhance algorithm performance by effectively reusing past experiences. Through the utilization of simulation studies, it can be demonstrated that the enhanced algorithm exhibits superior performance in comparison with the pre-existing method.

Protocol for a distributed smart building solution using semi-physical simulation.

Honeycomb is a distributed smart building system that is robust, flexible, and portable. Here, we present a protocol that uses semi-physical simulation to construct a Honeycomb prototype. We describe steps for software and hardware preparation, as well as the implementation of a video-based occupancy detection algorithm. Besides, we provide examples and scenarios of distributed applications, including node failure and recovery. We further provide guidance on data visualization and analysis to facilitate the design of distributed applications for smart buildings. For complete details on the use and execution of this protocol, please refer to Xing et al.1.

LQR Control and Optimization for Trajectory Tracking of Biomimetic Robotic Fish Based on Unreal Engine.

A realistic and visible dynamic simulation platform can significantly facilitate research on underwater robots. This paper uses the Unreal Engine to generate a scene that resembles real ocean environments, before building a visual dynamic simulation platform in conjunction with the Air-Sim system. On this basis, the trajectory tracking of a biomimetic robotic fish is simulated and assessed. More specifically, we propose a particle swarm optimization algorithm-based control strategy to optimize the discrete linear quadratic regulator controller for the trajectory tracking problem, as well as tracking and controlling discrete trajectories with misaligned time series through introducing a dynamic time warping algorithm. Simulation analyses of the biomimetic robotic fish following a straight line, a circular curve without mutation, and a four-leaf clover curve with mutation are carried out. The obtained results verify the feasibility and effectiveness of the proposed control strategy.

Distributed Relative Localization Algorithms for Multi-Robot Networks: A Survey.

For a network of robots working in a specific environment, relative localization among robots is the basis for accomplishing various upper-level tasks. To avoid the latency and fragility of long-range or multi-hop communication, distributed relative localization algorithms, in which robots take local measurements and calculate localizations and poses relative to their neighbors distributively, are highly desired. Distributed relative localization has the advantages of a low communication burden and better system robustness but encounters challenges in the distributed algorithm design, communication protocol design, local network organization, etc. This paper presents a detailed survey of the key methodologies designed for distributed relative localization for robot networks. We classify the distributed localization algorithms regarding to the types of measurements, i.e., distance-based, bearing-based, and multiple-measurement-fusion-based. The detailed design methodologies, advantages, drawbacks, and application scenarios of different distributed localization algorithms are introduced and summarized. Then, the research works that support distributed localization, including local network organization, communication efficiency, and the robustness of distributed localization algorithms, are surveyed. Finally, popular simulation platforms are summarized and compared in order to facilitate future research and experiments on distributed relative localization algorithms.

Honeycomb: An open-source distributed system for smart buildings.

Restricted by the hierarchical and centralized system architecture, smart buildings face challenges such as limited adaptability and robustness, single application functionalities, and complex configurations. To address the above shortcomings, we learn from the activity patterns of natural bee swarms and propose Honeycomb, an open-source smart-building solution with fully distributed architecture. Honeycomb is a robust, flexible smart-building solution without any central server or global leader. An asynchronous leaderless spanning tree-based communication pattern is developed to generate and maintain the communication topology of Honeycomb in real time. Benefiting from this communication pattern, Honeycomb has plug-and-play ability. Various distributed applications are designed for building operating tasks and are deployed in a real Honeycomb prototype. The prototype demonstrates significant energy efficiency improvement from the control of the heating, ventilation, and air conditioning (HVAC) system with video-based occupancy information. Feedback on our Honeycomb prototype through questionnaires of users shows high acceptance of the controlled indoor environment.

Research on Classification of COVID-19 Chest X-Ray Image Modal Feature Fusion Based on Deep Learning.

Most detection methods of coronavirus disease 2019 (COVID-19) use classic image classification models, which have problems of low recognition accuracy and inaccurate capture of modal features when detecting chest X-rays of COVID-19. This study proposes a COVID-19 detection method based on image modal feature fusion. This method first performs small-sample enhancement processing on chest X-rays, such as rotation, translation, and random transformation. Five classic pretraining models are used when extracting modal features. A global average pooling layer reduces training parameters and prevents overfitting. The model is trained and fine-tuned, the machine learning evaluation standard is used to evaluate the model, and the receiver operating characteristic (ROC) curve is drawn. Experiments show that compared with the classic model, the classification method in this study can more effectively detect COVID-19 image modal information, and it achieves the expected effect of accurately detecting cases.

Epidemic spreading in heterogeneous networks with recurrent mobility patterns.

Much recent research has shown that network structure and human mobility have great influences on epidemic spreading. In this paper, we propose a discrete-time Markov chain method to model susceptible-infected-susceptible epidemic dynamics in heterogeneous networks. There are two types of locations, residences and common places, for which different infection mechanisms are adopted. We also give theoretical results about the impacts of important factors, such as mobility probability and isolation, on epidemic threshold. Numerical simulations are conducted, and experimental results support our analysis. In addition, we find that the dominations of different types of residences might reverse when mobility probability varies for some networks. In summary, the findings are helpful for policy making to prevent the spreading of epidemics.

Epidemic in networked population with recurrent mobility pattern.

The novel Coronavirus (COVID-19) has caused a global crisis and many governments have taken social measures, such as home quarantine and maintaining social distance. Many recent studies show that network structure and human mobility greatly influence the dynamics of epidemic spreading. In this paper, we utilize a discrete-time Markov chain approach and propose an epidemic model to describe virus propagation in the heterogeneous graph, which is used to represent individuals with intra social connections and mobility between individuals and common locations. There are two types of nodes, individuals and public places, and disease can spread by social contacts among individuals and people gathering in common areas. We give theoretical results about epidemic threshold and influence of isolation factor. Several numerical simulations are performed and experimental results further demonstrate the correctness of proposed model. Non-monotonic relationship between mobility possibility and epidemic threshold and differences between Erdös-Rényi and power-law social connections are revealed. In summary, our proposed approach and findings are helpful to analyse and prevent the epidemic spreading in networked population with recurrent mobility pattern.

Individual mobility promotes punishment in evolutionary public goods games.

In explaining the pressing issue in biology and social sciences how cooperation emerges in a population of self-interested individuals, researchers recently pay intensive attentions to the role altruistic punishment plays. However, as higher-order cooperators, survival of punishers is puzzling due to their extra cost in regulating norm violators. Previous works have highlighted the importance of individual mobility in promoting cooperation. Yet its effect on punishers remains to be explored. In this work we incorporate this feature into modeling the behavior of punishers, who are endowed with a choice between leaving current place or staying and punishing defectors. Results indicate that optimal mobility level of punishers is closely related to the cost of punishing. For considerably large cost, there exists medium tendency of migration which favors the survival of punishers. This holds for both the direct competition between punishers and defectors and the case where cooperators are involved, and can also be observed when various types of punishers with different mobility tendencies fight against defectors simultaneously. For cheap punishment, mobility does not provide with punishers more advantage even when they are initially rare. We hope our work provide more insight into understanding the role individual mobility plays in promoting public cooperation.

Cumulative Dynamics of Independent Information Spreading Behaviour: A Physical Perspective.

The popularization of information spreading in online social networks facilitates daily communication among people. Although much work has been done to study the effect of interactions among people on spreading, there is less work that considers the pattern of spreading behaviour when people independently make their decisions. By comparing microblogging, an important medium for information spreading, with the disordered spin glass system, we find that there exist interesting corresponding relationships between them. And the effect of aging can be observed in both systems. Based on the analogy with the Trap Model of spin glasses, we derive a model with a unified power-function form for the growth of independent spreading activities. Our model takes several key factors into consideration, including memory effect, the dynamics of human interest, and the fact that older messages are more difficult to discover. We validate our model by a real-world microblogging data set. Our work indicates that, other than various features, some invariable rules should be considered during spreading prediction. This work also contributes a useful methodology for studying human dynamics.

Impact of Repeated Exposures on Information Spreading in Social Networks.

Clustered structure of social networks provides the chances of repeated exposures to carriers with similar information. It is commonly believed that the impact of repeated exposures on the spreading of information is nontrivial. Does this effect increase the probability that an individual forwards a message in social networks? If so, to what extent does this effect influence people's decisions on whether or not to spread information? Based on a large-scale microblogging data set, which logs the message spreading processes and users' forwarding activities, we conduct a data-driven analysis to explore the answer to the above questions. The results show that an overwhelming majority of message samples are more probable to be forwarded under repeated exposures, compared to those under only a single exposure. For those message samples that cover various topics, we observe a relatively fixed, topic-independent multiplier of the willingness of spreading when repeated exposures occur, regardless of the differences in network structure. We believe that this finding reflects average people's intrinsic psychological gain under repeated stimuli. Hence, it makes sense that the gain is associated with personal response behavior, rather than network structure. Moreover, we find that the gain is robust against the change of message popularity. This finding supports that there exists a relatively fixed gain brought by repeated exposures. Based on the above findings, we propose a parsimonious model to predict the saturated numbers of forwarding activities of messages. Our work could contribute to better understandings of behavioral psychology and social media analytics.

A formal model for analyzing drug combination effects and its application in TNF-alpha-induced NFkappaB pathway.

BACKGROUND: Drug combination therapy is commonly used in clinical practice. Many methods including Bliss independence method have been proposed for drug combination design based on simulations models or experiments. Although Bliss independence method can help to solve the drug combination design problem when there are only a small number of combinations, as the number of combinations increases, it may not be scalable. Exploration of system structure becomes important to reduce the complexity of the design problem. RESULTS: In this paper, we deduced a mathematical model which can simplify the serial structure and parallel structure of biological pathway for synergy evaluation of drug combinations. We demonstrated in steady state the sign of the synergism assessment factor derivative of the original system can be predicted by the sign of its simplified system. In addition, we analyzed the influence of feedback structure on survival ratio of the serial structure. We provided a sufficient condition under which the combination effect could be maintained. Furthermore, we applied our method to find three synergistic drug combinations on tumor necrosis factor alpha-induced NFkappaB pathway and subsequently verified by the cell experiment. CONCLUSIONS: We identified several structural properties underlying the Bliss independence criterion, and developed a systematic simplification framework for drug combination design by combining simulation and system reaction network topology analysis. We hope that this work can provide insights to tackle the challenging problem of assessment of combinational drug therapy effect in a large scale signaling pathway. And hopefully in the future our method could be expanded to more general criteria.

A remark on "scalar equations for synchronous Boolean networks with biological applications" by C. Farrow, J. Heidel, J. Maloney, and J. Rogers.

The problem of finding all cycles in the exponentially growing state space of synchronous Boolean networks was studied in the paper by C. Farrow, J. Heidel, J. Maloney, and J. R. Scalar, "Equations for synchronous Boolean networks with biological applications," IEEE trans. Neural Networks, vol. 15, no. 2, pp. 348-354 Mar. 2004. No efficient algorithm was given to solve the problem. We show that even the determination of the number of fixed points (cycles of length 1) for monotone Boolean networks and the determination of the existence of fixed points for general Boolean networks are both strong NP-complete.