Complex Dynamic Environments Research Articles

A Task Allocation Strategy of the UAV Swarm Based on Multi-Discrete Wolf Pack Algorithm

With the continuous development of artificial intelligence, swarm control and other technologies, the application of Unmanned Aerial Vehicles (UAVs) in the battlefield is more and more extensive, and the UAV swarm is increasingly playing a prominent role in the future of warfare. How tasks are assigned in the dynamic and complex battlefield environment is very important. This paper proposes a task assignment model and its objective function based on dynamic information convergence. In order to resolve this multidimensional function, the Wolf Pack Algorithm (WPA) is selected as the alternative optimization algorithm. This is because its functional optimization of high-dimensional complex problems is better than other intelligent algorithms, and the fact that it is more suitable for UAV swarm task allocation scenarios. Based on the traditional WPA algorithm, this paper proposes a Multi-discrete Wolf Pack Algorithm (MDWPA) to solve the UAV task assignment problem in a complex environment through the discretization of wandering, calling, sieging behavior, and new individual supplement. Through Orthogonal Experiment Design (OED) and analysis of variance, the results show that MDWPA performs with better accuracy, robustness, and convergence rate and can effectively solve the task assignment problem of UAVs in a complex dynamic environment.

Engineering Pt heterogeneous catalysts for accelerated liquid–solid redox conversion in Li-S batteries

The shuttle effect caused by soluble lithium polysulfides (LiPSs) deteriorates multiphase transformation reaction kinetics of sulfur species, and gives rise to an unserviceable lithium-sulfur (Li-S) battery. Catalysis, as a process optimization approach, offers an option to eliminate the intrinsic issues. However, exploring and understanding the role of catalysts on electrode reaction remains critical bottlenecks, particularly as they are prone to continuous evolution under complex dynamic environment. Herein, platinum nanoparticles loaded on MXene nanosheets, as sulfur host, and the action of catalysts on the reaction process are investigated via ex-situ monitors upon solid–liquid–solid chemical transformation of sulfur species. These traces confirm that the high performance originates from electron transfer between catalysts and LiPSs, which lowers the nucleation barrier from liquid LiPSs to solid Li2S/Li2S2. Further, the accelerated liquid–solid conversion can alleviate the accumulation of LiPSs, and boost the reaction kinetics in Li-S batteries. The findings corroborate the electronic modulation between catalysts and LiPSs, which is a generalizable strategy to optimize energy conversion efficiency of Li-S batteries.

An Effective Dynamic Path Planning Approach for Mobile Robots Based on Ant Colony Fusion Dynamic Windows

To further improve the path planning of the mobile robot in complex dynamic environments, this paper proposes an enhanced hybrid algorithm by considering the excellent search capability of the ant colony optimization (ACO) for global paths and the advantages of the dynamic window approach (DWA) for local obstacle avoidance. Firstly, we establish a new dynamic environment model based on the motion characteristics of the obstacles. Secondly, we improve the traditional ACO from the pheromone update and heuristic function and then design a strategy to solve the deadlock problem. Considering the actual path requirements of the robot, a new path smoothing method is present. Finally, the robot modeled by DWA obtains navigation information from the global path, and we enhance its trajectory tracking capability and dynamic obstacle avoidance capability by improving the evaluation function. The simulation and experimental results show that our algorithm improves the robot’s navigation capability, search capability, and dynamic obstacle avoidance capability in unknown and complex dynamic environments.

Real-Time Ultrasound Doppler Tracking and Autonomous Navigation of a Miniature Helical Robot for Accelerating Thrombolysis in Dynamic Blood Flow.

Untethered small-scale robots offer great promise for medical applications in complex biological environments. However, challenges remain in the control and medical imaging of a robot for targeted delivery inside a living body, especially in flowing conditions (e.g., blood vessels). In this work, we report a strategy to autonomously navigate a miniature helical robot in dynamic blood flow under ultrasound Doppler imaging guidance. A magnetic torque and force-hybrid control approach is implemented, enabling the actuation of a millimeter-scale helical robot against blood flow under a rotating magnetic field with a controllable field gradient. Experimental results demonstrate that the robot (length 7.30 mm; diameter 2.15 mm) exhibits controlled navigation in vascular environments, including upstream and downstream navigation in flowing and pulsatile flowing blood with flow rates up to 24 mL/min (mean flow velocity: 14.15 mm/s). During navigation, the rotating robot-induced Doppler signals enable real-time localization and tracking in flowing and pulsatile flowing blood environments. Moreover, the robot can be selectively navigated along different paths by actively controlling the robot's orientation. We apply this autonomous strategy for localizing thrombus and accelerating thrombolysis rate. Compared with conventional tissue plasminogen activator (tPA) thrombolysis, the robot-enhanced shear stress and tPA convection near the clot-blood interface increase the unblocking and thrombolysis efficiency up to 4.8- and 3.5-fold, respectively. Such a medical imaging-guided navigation strategy provides simultaneous robot navigation and localization in complex dynamic biological environments, providing an intelligent approach toward real-time targeted delivery and diagnostic applications in vivo.

A Detection Model of the Complex Dynamic Traffic Environment for Unmanned Vehicles

It has always been an important and arduous task to detect the complex dynamic traffic environment, especially for unmanned driving. Although the existing advanced detection models have reached the speed requirements for detection, the detection accuracy needs to be further elevated to improve the unmanned driving’s safety. How to balance the accuracy and speed of detecting the complex dynamic traffic environment is still the primary problem to be solved for unmanned vehicles. Therefore, this article proposes a detection model of the complex dynamic traffic environment for unmanned vehicles by following the framework idea of YOLOv3. Firstly, we regard MobileNetv3 as the backbone and replace the traditional convolution with the depthwise separable convolution in the whole model to reduce the number of parameters and calculations. Secondly, in enhanced feature fusion layers, we perform the multi-scale fusion of four feature maps by the compress-and-expand module, the SPP module, and the cross-layer bidirectional module of feature fusion to improve the locating accuracy and reduce false detections. Thirdly, we add an IoU loss to improve the accuracy of model regression. Then, we employ the improved clustering algorithm to re-cluster anchor boxes, reducing the time overhead while improving the clustering accuracy. Finally, we compare the proposed model with other advanced detection models in the processed BDD dataset and the KITTI dataset. We verify that the mAP of the proposed model improves notably without loss of detection speed, the number of parameters and calculations decreases dramatically, and the proposed model exhibits a more superior performance.

Path planning and collision avoidance methods for distributed multi-robot systems in complex dynamic environments.

Multi-robot systems are experiencing increasing popularity in joint rescue, intelligent transportation, and other fields. However, path planning and navigation obstacle avoidance among multiple robots, as well as dynamic environments, raise significant challenges. We propose a distributed multi-mobile robot navigation and obstacle avoidance method in unknown environments. First, we propose a bidirectional alternating jump point search A* algorithm (BAJPSA*) to obtain the robot's global path in the prior environment and further improve the heuristic function to enhance efficiency. We construct a robot kinematic model based on the dynamic window approach (DWA), present an adaptive navigation strategy, and introduce a new path tracking evaluation function that improves path tracking accuracy and optimality. To strengthen the security of obstacle avoidance, we modify the decision rules and obstacle avoidance rules of the single robot and further improve the decision avoidance capability of multi-robot systems. Moreover, the mainstream prioritization method is used to coordinate the local dynamic path planning of our multi-robot systems to resolve collision conflicts, reducing the difficulty of obstacle avoidance and simplifying the algorithm. Experimental results show that this distributed multi-mobile robot motion planning method can provide better navigation and obstacle avoidance strategies in complex dynamic environments, which provides a technical reference in practical situations.

Construction of Novel Self-Adaptive Dynamic Window Approach Combined With Fuzzy Neural Network in Complex Dynamic Environments

The traditional Dynamic Window Approach (DWA) with constant weight values of the evaluation function leads to the inability of obstacle avoidance for the Automated Guided Vehicles (AGV) to perform obstacle avoidance and path planning in the complex environment. Effective avoidance of complex obstacles requires adaptive weight adjustment to address the evaluation function’s challenges. This paper proposes an adaptive DWA (ADWA), which introduces neural network training on the basis of the Mamdani DWA (MDWA). Firstly, the Mamdani type fuzzy controller is designed, and then the adaptive neuro-fuzzy controller is obtained by neural network training. Then, experiments are carried out through the MATLAB simulation environment. The simulation experiment results show that the improved DWA compared to traditional DWA can make the AGV pass the obstacle environment with a better trajectory and reduce the time. The improved DWA improves the autonomous obstacle avoidance capability of AGVs, which not only perfectly fits our task requirements, but also has apparent scientific and practical significance in developing AGV autonomous obstacle avoidance technology.

A Mobile Robot Path Planning Algorithm Based on Improved A* Algorithm and Dynamic Window Approach

The traditional A* algorithm has several problems in practical applications, such as many path turning points, redundant nodes, and long running time. it is sometimes impossible to plan the theoretical optimal route. To solve the above problem, this paper presents an optimized A* algorithm, the adaptive adjustment step algorithm and the three-time Bezier curve are used to solve the problems of many turning points, large turning angles, and long running time in the search path. Moreover, aiming at the path planning problem of mobile robots facing dynamic obstacle interference in complex environments, an algorithm that integrates the improved A* algorithm with the dynamic window method is proposed, which not only solves the shortcomings of the A* algorithm in which the dynamic obstacles cannot be avoided, but also prevents the mobile robot from falling into local optimization. The results show that the fusion algorithm of the improved A* algorithm and the dynamic window method with the traditional A* algorithm reduces the number of turns by 50% and the path length by 3.62% compared with the original algorithm. In the same environment, compared with the traditional algorithm, the hybrid algorithm in this paper reduces the average time consumption by 10.27%, the number of path inflection points by 57.14%, and the accuracy is higher than 33.33%, which is more effective in complex dynamic environments.

Clustered Autoencoded Variational Inverse Reinforcement Learning

Abstract Variational Auto-Encoder (VAE) is a handy and computationally friendly Bayesian tool for Inverse Reinforcement Learning (IRL) problems, with the native setting of absent reward functions in a Markov Decision Process (MDP). However, recent works mainly deal with single reward, which turn out to be insufficient for complex dynamic environments with multiple demonstrators of various characteristics (hence multiple reward functions). This paper extends the dimensionality of reward (from ℝ to ℝ K ) by incorporating a latent embedding and clustering step on top of a scalable Bayesian IRL model, which enhances her applicability to multi-reward scenarios. We introduce our method, Clustered Autoencoded Variational Inverse Reinforcement Learning (CAVIRL), which is able to approximate multiple posterior reward functions and learn the corresponding policies for experts of various characteristics and skills. As a by-product, the proposed model also thrives to determine the number of clusters K on her own, as opposed to the competing multi-reward imitation learning models that require K to be prespecified. We trained the proposed model within a grid world with multiple types of players, where we achieved 100% correctness in determining the number of players’ types and 80%-83.9% match between the model-learned policies and the players’ demonstrations from the data.

Modeling cell size control under dynamic environments

In nature, cells face changes in environmental conditions that can modify their growth rate. In these dynamic environments, recent experiments found changes in cell size regulation. Currently, there are few clues about the origin of these changes in cell size. In this work, we model cell division as a stochastic process that occurs at a rate proportional to the size. We propose that this rate is zero if the cell is smaller than a minimum size. We show how this model predicts some of the properties found in cell size regulation. For example, among our predictions, we found that the mean cell size is an exponential function of the growth rate under steady conditions. We predict that cells become smaller and how the division strategy changes during dynamic nutrient depletion. Finally, we use the model to predict cell regulation in an arbitrary complex dynamic environment.

ME‐MADDPG: An efficient learning‐based motion planning method for multiple agents in complex environments

Developing efficient motion policies for multiagents is a challenge in a decentralized dynamic situation, where each agent plans its own paths without knowing the policies of the other agents involved. This paper presents an efficient learning-based motion planning method for multiagent systems. It adopts the framework of multiagent deep deterministic policy gradient (MADDPG) to directly map partially observed information to motion commands for multiple agents. To improve the efficiency of MADDPG in sample utilization, so as to train more brilliant agents that can adapt to more complex environments, a strategy named mixed experience (ME) is introduced to MADDPG, and this has led to our proposed ME-MADDPG algorithm. The novel ME strategy can be embodied into three specific mechanisms: (1) an artificial potential field-based sample generator to produce high-quality samples in the early training stage; (2) a dynamic mixed sampling strategy to mix the training data from different sources with a variable proportion; (3) a delayed learning skill to stabilize the training of the multiple agents. A series of experiments have been conducted to verify the performance of the proposed ME-MADDPG algorithm, and it has been demonstrated that, compared with MADDPG, the proposed algorithm can significantly improve the convergence speed and convergence effect in the training process, and it has also shown better efficiency and better adaptability in complex dynamic environments while it is used for multiagent motion planning applications.

Social Robot Navigation Tasks: Combining Machine Learning Techniques and Social Force Model.

Social robot navigation in public spaces, buildings or private houses is a difficult problem that is not well solved due to environmental constraints (buildings, static objects etc.), pedestrians and other mobile vehicles. Moreover, robots have to move in a human-aware manner—that is, robots have to navigate in such a way that people feel safe and comfortable. In this work, we present two navigation tasks, social robot navigation and robot accompaniment, which combine machine learning techniques with the Social Force Model (SFM) allowing human-aware social navigation. The robots in both approaches use data from different sensors to capture the environment knowledge as well as information from pedestrian motion. The two navigation tasks make use of the SFM, which is a general framework in which human motion behaviors can be expressed through a set of functions depending on the pedestrians’ relative and absolute positions and velocities. Additionally, in both social navigation tasks, the robot’s motion behavior is learned using machine learning techniques: in the first case using supervised deep learning techniques and, in the second case, using Reinforcement Learning (RL). The machine learning techniques are combined with the SFM to create navigation models that behave in a social manner when the robot is navigating in an environment with pedestrians or accompanying a person. The validation of the systems was performed with a large set of simulations and real-life experiments with a new humanoid robot denominated IVO and with an aerial robot. The experiments show that the combination of SFM and machine learning can solve human-aware robot navigation in complex dynamic environments.

Model predictive control for reusable space launcher guidance improvement

This work aims to demonstrate the benefits and limitations of an on-board Guidance for reusable launch vehicles, as well as to tradeoff different Model Predictive Control (MPC)-based Guidance and Control (G&C) architectures, exploiting, in particular, recent advances on successive convexification algorithms for optimization problems. Leading space agencies and private companies are investing on the development of reusable space launchers to reduce the cost to access the space. Indeed, that cost is one of the major deterrents in space exploration and space utilization. Reusability is, therefore, the unanimous solution to lower costs, and get a reliable and fast space access. Among many technological enhancements, the guidance, navigation, and control plays a crucial role: precise pinpoint landing capabilities or mid-air recovery, in fact, are mandatory. Indeed, the capability for generating re-optimized guidance trajectories on-board in real-time based on current flight conditions promises to improve the system performance, allows for fault tolerance capabilities, and reduces mission preparation costs. The work focuses especially on the implementation of a successive convexification Model Predictive Control guidance algorithm which solves the 6 Degree-of-Freedom (DoF) Powered Descent Guidance problem (PDG). The novelty of that work is applying a model predictive-based technique to a complex dynamic environment, trading off different solutions to the problem and relying on results obtained by using an industrial simulation framework. The robustness of the proposed approach is tested in several operative scenarios and the feasibility of real-time implementation is studied. For what concerns the trajectory optimization routine, the formulation of the problem, while initially being non-convex, is convexified. This is performed by implementing a successive convexification algorithm, which obtains a sub-optimal solution of the original problem in a fraction of the time required by a global optimizer, by solving a Second Order Cone Programming (SOCP) problem. This method allows coping with different kinds of dynamics nonlinearities, as well as cost functions and constraints. By presenting the approach and critically discussing the obtained results, the work provides an overview of the different methodologies available in the literature and assesses the limits of those approaches when applied to highly nonlinear scenarios, with large dispersions of uncertain parameters, as it is the case of reusable launch vehicles.

Using a probabilistic neural network to assess the state of information security

Modern cyber-physical systems incorporate a large volume of various information exchange processes that frame a complex dynamic environment with multiple tasks to be undertaken for monitoring, diagnosing, predicting and detecting hazardous and safe states of system objects, which entails a need to ensure information security of devices and elements in the Industrial Internet and the Internet of Things. The paper proposes a method for monitoring the state of information security based on processing information gained after side-channel processing. The state is evaluated based on time series describing the values of parameters obtained at successive points in time, related, for example, to power consumption, CPU utilization, internal memory usage, and intensity of messaging used to monitor abnormal states.

Modeling Context Aware Dynamic Trust Using Hidden Markov Model

Modeling trust in complex dynamic environments is an important yet challenging issue since an intelligent agent may strategically change its behavior to maximize its profits. In thispaper, we propose a context aware trust model to predict dynamic trust by using a Hidden Markov Model (HMM) to model an agent's interactions. Although HMMs have already been applied in the past to model an agent's dynamic behavior to greatly improve the traditional static probabilistic trust approaches, most HMM based trust models only focus on outcomes of the past interactions without considering interaction context, which we believe, reflects immensely on the dynamic behavior or intent of an agent. Interaction contextual information is comprehensively studied and integrated into the model to more precisely approximate an agent's dynamic behavior. Evaluation using real auction data and synthetic data demonstrates the efficacy of our approach in comparison with previous state-of-the-art trust mechanisms.

LiDAR-Based SLAM under Semantic Constraints in Dynamic Environments

Facing the realistic demands of the application environment of robots, the application of simultaneous localisation and mapping (SLAM) has gradually moved from static environments to complex dynamic environments, while traditional SLAM methods usually result in pose estimation deviations caused by errors in data association due to the interference of dynamic elements in the environment. This problem is effectively solved in the present study by proposing a SLAM approach based on light detection and ranging (LiDAR) under semantic constraints in dynamic environments. Four main modules are used for the projection of point cloud data, semantic segmentation, dynamic element screening, and semantic map construction. A LiDAR point cloud semantic segmentation network SANet based on a spatial attention mechanism is proposed, which significantly improves the real-time performance and accuracy of point cloud semantic segmentation. A dynamic element selection algorithm is designed and used with prior knowledge to significantly reduce the pose estimation deviations caused by SLAM dynamic elements. The results of experiments conducted on the public datasets SemanticKITTI, KITTI, and SemanticPOSS show that the accuracy and robustness of the proposed approach are significantly improved.

Transfers and orbital maintenance of spatial retrograde orbits for Phobos exploration

Quasi-satellite orbits (QSOs) are stable retrograde orbits in the restricted three-body problem that have gained attention as a viable candidate for future deep-space missions towards remote planetary satellites. JAXA’s robotic sample return mission MMX will utilize QSOs to perform scientific observations of the Martian moon Phobos before landing on its surface and attempt sample retrieval. The complex dynamical environment around Phobos makes the proximity operations of MMX quite challenging and requires novel and sophisticated techniques for maintaining and transferring between different quasi-satellite orbits. The present paper explores the application of invariant manifolds of unstable retrograde orbits to design transfer trajectories around Phobos. Starting from the equations of the Circular Hill Problem with ellipsoidal Phobos, we first compute families of three-dimensional QSOs using out-of-plane bifurcations near planar orbits. The feasibility of using unstable family members as staging orbits between high-altitude and low-altitude QSOs is later assessed. The final candidates are ranked based on MMX scientific requirements, transfer analyses, and station-keeping costs. It is found that intermediate 3D-QSOs can be maintained with as little as 1m/s per month. Furthermore, it is discovered that transfer from high-altitude QSOs to low-altitude QSOs can be executed with a total ΔV of less than 40m/s and total time of flight of less than 5 days.

A Semantic SLAM System for Catadioptric Panoramic Cameras in Dynamic Environments.

When a traditional visual SLAM system works in a dynamic environment, it will be disturbed by dynamic objects and perform poorly. In order to overcome the interference of dynamic objects, we propose a semantic SLAM system for catadioptric panoramic cameras in dynamic environments. A real-time instance segmentation network is used to detect potential moving targets in the panoramic image. In order to find the real dynamic targets, potential moving targets are verified according to the sphere’s epipolar constraints. Then, when extracting feature points, the dynamic objects in the panoramic image are masked. Only static feature points are used to estimate the pose of the panoramic camera, so as to improve the accuracy of pose estimation. In order to verify the performance of our system, experiments were conducted on public data sets. The experiments showed that in a highly dynamic environment, the accuracy of our system is significantly better than traditional algorithms. By calculating the RMSE of the absolute trajectory error, we found that our system performed up to 96.3% better than traditional SLAM. Our catadioptric panoramic camera semantic SLAM system has higher accuracy and robustness in complex dynamic environments.

Resource-rational Models of Human Goal Pursuit.

Goal-directed behavior is a deeply important part of human psychology. People constantly set goals for themselves and pursue them in many domains of life. In this paper, we develop computational models that characterize how humans pursue goals in a complex dynamic environment and test how well they describe human behavior in an experiment. Our models are motivated by the principle of resource rationality and draw upon psychological insights about people's limited attention and planning capacities. We find that human goal pursuit is qualitatively different and substantially less efficient than optimal goal pursuit in our simulated environment. Models of goal pursuit based on the principle of resource rationality capture human behavior better than both a model of optimal goal pursuit and heuristics that are not resource-rational. We conclude that the way humans pursue goals is shaped by the need to achieve goals effectively as well as cognitive costs and constraints on planning and attention. Our findings are an important step toward understanding humans' goal pursuit as cognitive limitations play a crucial role in shaping people's goal-directedbehavior.

Gravity and Known Size Calibrate Visual Information to Time Parabolic Trajectories.

Catching a ball in a parabolic flight is a complex task in which the time and area of interception are strongly coupled, making interception possible for a short period. Although this makes the estimation of time-to-contact (TTC) from visual information in parabolic trajectories very useful, previous attempts to explain our precision in interceptive tasks circumvent the need to estimate TTC to guide our action. Obtaining TTC from optical variables alone in parabolic trajectories would imply very complex transformations from 2D retinal images to a 3D layout. We propose based on previous work and show by using simulations that exploiting prior distributions of gravity and known physical size makes these transformations much simpler, enabling predictive capacities from minimal early visual information. Optical information is inherently ambiguous, and therefore, it is necessary to explain how these prior distributions generate predictions. Here is where the role of prior information comes into play: it could help to interpret and calibrate visual information to yield meaningful predictions of the remaining TTC. The objective of this work is: (1) to describe the primary sources of information available to the observer in parabolic trajectories; (2) unveil how prior information can be used to disambiguate the sources of visual information within a Bayesian encoding-decoding framework; (3) show that such predictions might be robust against complex dynamic environments; and (4) indicate future lines of research to scrutinize the role of prior knowledge calibrating visual information and prediction for action control.