An ocean current-guided RRT∗ algorithm for Multi-USV path planning

An Adaptive Multi-Population Cooperative Whale Optimization Algorithm for global optimization and 3D UAV path planning

Rolling-horizon genetic algorithm for adaptive path planning in hazardous environments

Acute ischemic stroke is a leading cause of morbidity and mortality requiring timely intervention. Mechanical thrombectomy is an increasingly prevalent technique for removal of the clot burden in patients with acute ischemic stroke with a targetable vessel, enabling treatment within 24 hours of symptom onset. In the post-mechanical thrombectomy period, patients must be monitored closely for complications and in preparation for secondary prevention therapy. Patients typically undergo serial follow-up imaging examinations with multiple modalities, including newer techniques such as dual-energy CT. The radiologist should be aware of the expected findings in the post-mechanical thrombectomy period, such as contrast material staining, and be able to distinguish these from complications. Post-mechanical thrombectomy complications can arise anywhere along the instrumentation path, involving both intracranial and extracranial findings. A conceptual understanding of the clinical implications of various complications is presented to help radiologists appropriately contextualize their findings at follow-up imaging. ©RSNA, 2025 Supplemental material is available for this article.

Background: To address the problem of high energy consumption of self-driving electric vehicles when following a planned trajectory, constraints are added to path planning and speed planning respectively. Methods: Given the limitations of the existing path planning algorithms in terms of search efficiency and path length, this study introduces an innovative and improved strategy in the horizontal dimension. Based on the cost function of the distance between sampling points, this strategy aims to improve the search efficiency of the dynamic planning algorithm and reduce the search path length. Furthermore, the smoothness of the path is optimized to suit the actual driving conditions by applying a quadratic programming algorithm. An energy consumption model for pure electric vehicles is established in the vertical dimension, effectively constraining energy use during speed dynamic planning to reduce consumption while driving. Finally, the smoothness of speed planning is improved using a quadratic programming algorithm. Results: The results of simulation experiments show that compared with traditional methods, the proposed algorithm achieves a substantial improvement in path length reduction of 5.8%, average curvature reduction of 31.6%, and average energy consumption reduction of 2.04% in static and dynamic obstacle avoidance environments. Conclusion: The results show that the improved dynamic planning algorithm proposed in this study is significantly optimized in terms of mean path length, mean curvature, and energy consumption. Moreover, the proposed algorithm can meet the requirements of energy efficiency of vehicle driving.

High-performance multi-agent path finding in high-obstacle-density and large-size maps

This paper proposes a Hybrid Complete Coverage Path Planning (HCCPP) algorithm to enhance the efficiency and smoothness of suspended mowers operating in convex polygonal fields. Combining straight-in, nested, and outward-spiral strategies, it optimizes internal and boundary coverage while using Hybrid A* and Bézier curves for smooth transitions. The simulation experiment uses a suspended lawn mower (Dongfanghong LX804 rear lawn mower) as the test platform, with parameters: dimensions of 6.50 m × 2.17 m × 2.87 m, working width of 2.5 m, minimum turning radius of 6.2 m, and minimum row spacing of 12.5 m. The simulation experiment of running 5 times for each of the 3 plots shows that HCCPP achieves >99.7% coverage, <5.4% overlap, 4–9% shorter paths, and lower curvature variation, outperforming traditional methods and offering an efficient solution for autonomous agricultural path planning.

In post-disaster search and rescue scenarios, robotic path planning must operate in unpredictable, dynamic environments where conventional coverage path planning (CPP) algorithms often struggle to adapt. To address this challenge, we propose an intelligent path planning algorithm called PERM-QN (Q-learning with priority experience replay and memory network), designed for energy-aware, complete area coverage in uncertain terrains. PERM-QN integrates a dynamic weight reward function, priority experience replay, and a memory network to enable efficient, comprehensive exploration in complex, obstacle-laden environments. The dynamic weight reward function adaptively balances coverage, path length, and energy consumption across different phases of operation, and the priority experience replay mechanism accelerates learning convergence by focusing on high-value past experiences. Finally, the memory network expedites route planning in regions with similar terrain, reducing redundant exploration. Experiments in simulated post-disaster environments of varying complexity demonstrate that PERM-QN achieves more efficient and comprehensive exploration than traditional methods while maintaining robust performance. These findings highlight PERM-QN as an effective path planning solution for robotic search in complex, dynamic environments.

The A* algorithm plays an important role in global path planning for robots, but it faces challenges such as redundant nodes and large search spaces. This paper proposes the Obstacle Density-based Dynamic Exponential A* (ODDEA*) algorithm. The ODDEA* algorithm adjusts the weights of the heuristic function based on the density of the surrounding obstacles. It uses the improved heuristic function to guide the robot toward areas with low obstacle density, employing a local dynamic penalty. The computational experiments compare the proposed ODDEA* algorithm with the Theta*, A*, and BA* algorithms, involving small-size (20×20), medium-size (40×40), and large-size (60×60) grid maps, as well as 50 random medium-size maps. The proposed ODDEA* algorithm uses fewer expanded nodes and less planning time than the other algorithms. Compared with the A* algorithm, it achieves 46.96% of the planning time and 20.33% of the search space on the three fixed grid maps.

There has been significant interest in solving robot path planning problems using fuzzy logic-based methods. Recently, the Genetic Algorithm-based Hierarchical Interval Type-2 Fuzzy (GA-HIT2F) system has been introduced as a novel planner in this domain. However, this method lacks adaptability to changes in target points, and insufficient flexibility can lead to planning failures in local minimum traps, making it difficult to apply to complex scenarios. In this paper, we identify the limitations of the original GA-HIT2F approach and propose an enhanced Q-Learning-aided Adaptive Hierarchical Interval Type-2 Fuzzy (QL-HIT2F) algorithm for path planning. The proposed planner incorporates reinforcement learning to improve a robot’s capability to avoid collisions with special obstacles. Additionally, the average obstacle orientation (AOO) is introduced to optimize the robot’s angular adjustments. Two supplementary robot parameters are integrated into the reinforcement learning action space, along with fuzzy membership parameters. The training process also introduces the concepts of meta-map and sub-training. Simulation results from a series of path planning experiments validate the feasibility and effectiveness of the proposed QL-HIT2F approach.

Precise and contactless manipulation of microstructures remains a major challenge in microassembly and lab-on-chip systems. Here, we introduced a Robotic Acoustic Streaming Tweezers (RAST) system that used an MEMS-based gigahertz resonator to generate localized acoustic streaming for programmable 2D and 3D manipulation of microgels. By tuning the power and height of the resonator, RAST achieved versatile, gentle operations of microgels. Our results demonstrated that the low-power enabled precise 2D translation and planar assembly of linear, layered, and patterned structures; intermediate power induced controlled flipping and rotation for orientation-specific alignment; and high power lifted microgels into 3D vortex traps, enabling accurate spatial nesting along guiding scaffolds. By integrating computer vision and path-planning algorithms, the system also enabled the high-throughput assembly of multiple microgels without human intervention. The RAST system offered a mild, label-free, and powerful strategy for complex microscale construction, providing a robust and scalable new solution for microrobots, biomedical devices, and advanced microfabrication technologies.

ABSTRACT To improve the auxiliary role of agricultural robots in agricultural farming, based on the screw theory, this paper studies the description of the position and posture coordinates of agricultural robots in two-dimensional and three-dimensional spaces. This paper then establishes the coordinate system of the manipulator in accordance with the regulations for establishing the coordinate system of agricultural robots. It uses mathematical formulas to derive and calculate the solution process of forward kinematics and inverse kinematics in detail, providing sample data for machine learning inverse kinematics algorithms. In addition, according to the control needs, this paper simplified the structure of the crawler mobile platform in the coordinate system. It established a kinematic model, which serves as the basis for the path-planning algorithm. Finally, this paper combines experimental research to verify the agricultural assisted agricultural robots developed in this paper. The experimental research results confirm the feasibility of this method.

The mainstream approach employing light detection and ranging (LiDAR) estimates the self-position of mobile robot by matching the point cloud acquired during navigation with that recorded in advance, in order to autonomously navigate to the goal point. However, this method is problematic in that it is vulnerable to environmental changes and that much effort and expenses are required to construct and update the point cloud map. Thus, in this paper, we propose an autonomous navigation method that does not require constructing a point cloud map by visiting the site in advance and is robust against environmental changes. The proposed method carries out autonomous navigation by using RTK-GNSS, and deep-learning algorithm of semantic segmentation and YOLO, A * algorithm for path planning, and pure pursuit algorithm for path navigation. Furthermore, obstacle avoidance is carried out using semantic segmentation, YOLO, and kernel density estimation. We conducted a navigation experiment, in which a 300 m section was autonomously navigated, thus verifying the validity of proposed method.

In mobile robotics, path planning enables autonomous navigation to specified destinations. However, complex terrain can lead to excessive tilting or even overturning, compromising stability and safety. Traditional path-planning algorithms often fail to fully account for dynamic terrain variations and robot motion constraints. To address these limitations, this paper proposes the novel dual-layer Hybrid-A* algorithm, enhanced with dynamic phase windows. This approach represents a significant innovation by integrating real-time feedback mechanisms and adaptive adjustments to phase windows, enabling continuous path refinement in response to both environmental changes and robot motion limitations. The guidance layer introduces a bicubic interpolation-based super-resolution technique to refine elevation maps, offering more accurate posture estimation. In the planning layer, we propose the dynamic use of multiple cost functions, an adaptive expansion radius, pruning strategies, and a phase-window activation mechanism, effectively addressing the computational challenges posed by large search spaces. The integration of these strategies allows the algorithm to outperform traditional methods, particularly in unstructured environments with complex terrain. Experimental results demonstrate the effectiveness of the proposed method in generating optimized paths that satisfy robot motion constraints, ensuring both efficiency and safety in real-world applications.

Digital microfluidic biochips (DMFBs) find extensive applications in biochemical experiments, medical diagnostics, and safety-critical domains, with their reliability dependent on efficient online testing technologies. However, traditional random search algorithms suffer from slow convergence and susceptibility to local optima under complex fluidic constraints. This paper proposes a hybrid optimization method based on priority strategy and an improved sparrow search algorithm for DMFB online test path planning. At the algorithmic level, the improved sparrow search algorithm incorporates three main components: tent chaotic mapping for population initialization, cosine adaptive weights together with Elite Opposition-based Learning (EOBL) to balance global exploration and local exploitation, and a Gaussian perturbation mechanism for fine-grained refinement of promising solutions. Concurrently, this paper proposes an intelligent rescue strategy that integrates global graph-theoretic pathfinding, local greedy heuristics, and space–time constraint verification to establish a closed-loop decision-making system. The experimental results show that the proposed algorithm is efficient. On the standard 7 × 7–15 × 15 DMFB benchmark chips, the shortest offline test path length obtained by the algorithm is equal to the length of the Euler path, indicating that, for these regular layouts, the shortest test path has reached the known optimal value. In both offline and online testing, the shortest paths found by the proposed method are better than or equal to those of existing mainstream algorithms. In particular, for the 15 × 15 chip under online testing, the proposed method reduces the path length from 543 and 471 to 446 compared with the IPSO and IACA algorithms, respectively, and reduces the standard deviation by 53.14% and 39.4% compared with IGWO in offline and online testing.

ABSTRACT The demand for cleaner, more efficient, and self-sufficient indoor spaces is rapidly increasing. Traditional manual cleaning methods are inconsistent and time-consuming, while most commercial robotic vacuum cleaners follow rigid, pre-set routes and lack intelligent dirt detection. However, advancements at the intersection of robotics, IoT, and computer vision are paving the way for unsupervised cleaning systems capable of sensing and responding to dirt in real time. This survey focuses on the development of a smart robotic arm equipped with a broom and integrated IoT-based computer vision for dust detection. We comprehensively review dust-cleaning robotic arms, computer vision techniques for dirt detection, remote monitoring via IoT, and autonomous indoor navigation. The evolution of cleaning robotic arms is traced from basic mechanical designs to sophisticated, vision-based systems that can identify dirt, utilize advanced algorithms for path optimization, and target cleaning with intelligent sweeping mechanisms. Research highlights the effectiveness of computer vision in recognizing dirt patterns, while IoT technologies enable real-time remote control and monitoring. Path-planning algorithms like A* further enhance navigation in complex environments. Although challenges such as lighting conditions, reflective surfaces, and mechanical constraints persist, the integration of these technologies holds significant promise for achieving truly autonomous and efficient indoor cleaning. This survey synthesizes key research trends and technological innovations that directly inform the design and implementation of the advanced automated floor-cleaning system developed for the capstone project.

In order to solve the problems of slow convergence speed, insufficient accuracy, and easily falling into the local optimum of the traditional whale optimization algorithm (WOA) in mobile robot path planning, an improved whale optimization algorithm (IWOA) combined with the bird navigation mechanism was proposed. Specific improvement measures include using logical chaos mapping to initialize the population to enhance the randomness and diversity of the initial solution, designing a nonlinear convergence factor to prevent the algorithm from prematurely entering the shrinking surround phase and extending the global search time, introducing an adaptive spiral shape constant to dynamically adjust the search range to balance exploration and development capabilities, optimizing the individual update strategy in combination with the bird navigation mechanism, and optimizing the algorithm through companion position information, thereby improving the stability and convergence speed of the algorithm. Path planning simulations were performed on 30 × 30 and 50 × 50 grid maps. The results show that compared with WOA, MSWOA, and GA, in the 30 × 30 map, the path length of IWOA is shortened by 3.23%, 7.16%, and 6.49%, respectively; in the 50 × 50 map, the path length is shortened by 4.88%, 4.53%, and 28.37%, respectively. This study shows that IWOA has significant advantages in the accuracy and efficiency of path planning, which verifies its feasibility and superiority.

Number of vehicles in metropolitan areas is rapidly increasing, leading to worsening traffic congestion. There is an urgent need for the implementation of effective vehicle routing planning (VRP) to increase road traffic efficiency. However, the existing path planning algorithms focus primarily on the optimization of single vehicles, neglecting the correlations in routing demands and the collective behavior of vehicles. To address these issues, this paper proposes evolutionary game-based multivehicle route planning (EG-MVRP). By constructing a vehicle grouping model and applying evolutionary game theory, we analyze the interest conflicts and coordination requirements among different vehicle groups, which leads to the formulation of an optimized cooperative path planning strategy for multiple vehicles. The experimental results demonstrate that EG-MVRP significantly enhances the efficiency of intragroup path planning, alleviates local congestion, and improves the overall operational efficiency of the traffic network by minimizing excessive competition among multiple vehicle groups on the same road sections. In addition, the proposed method offers clear advantages in reducing travel time and fuel consumption. The research presented in this paper offers novel ideas and methods for future traffic management and planning, offering significant practical application value.

Abstract In this paper, we propose a novel dynamic path planning method for multiple mobile robots, named CBS-TLTD3. The method combines the Conflict-Based Search (CBS) algorithm for global path planning and enhances the obstacle avoidance capability of the Twin Delayed Deep Deterministic Policy Gradient (TD3) model through transfer learning, thereby developing the TLTD3 model. Additionally, we introduce an intermediate planner to coordinate these two methods, enabling continuous path planning. This integrated approach enables real-time conflict detection and resolution, ensuring stable and continuous global path planning. Furthermore, the effectiveness of this integrated method is validated through simulations and real-world experiments, providing valuable insights and solutions for multi-robot path planning applications.

The growing adoption of Unmanned Surface Vessels (USVs) in commercial and defense domains raises challenges for safe navigation and strict adherence to the International Regulations for Preventing Collisions at Sea (COLREGs). This paper presents the implementation and field validation of three collision-avoidance approaches on a real USV: (i) behavior-based, (ii) a modified Velocity Obstacles (VO) algorithm, and (iii) a modified A* path-planning algorithm. Field trials in Guanabara Bay (Brazil) show that the behavior-based algorithm achieved the best balance between safety and efficiency, maintaining a safe mean Closest Point of Approach (30.0 m) while minimizing operational penalties: shortest total distance (179.4 m average), lowest mission completion time (174.7 s average), and smallest trajectory deviation (27.2% average increase). The VO algorithm operated with reduced safety margins (13.0 m average CPA) at the expense of larger detours (37.6% average distance increase), while the modified A* maintained equivalent safety (30.0 m average CPA) but produced the largest deviations (46.5% average increase). The trade-off analysis reveals that algorithm selection depends on operational priorities between safety margins and route efficiency.