Towards Industrial Scenario Lane Detection: Vision-Based AGV Navigation Methods

Automated Guided Vehicles (AGV) have received a lot of attention in recent years in terms of both hardware and software research. Nowadays, the AGV offers more adaptable and effective industrial and transportation system solutions. An AGV’s navigation technique is essential to its operation. The decision to use AGV navigation is not straightforward, even if it appears appropriate and sufficient. This paper surveys the navigation approaches applied to AGV in the past five years of published academic research. In doing so, this work responds to three related questions: 1) are the AGV’s classical navigation techniques still relevant to the current application area?; 2) are heuristic navigation techniques themselves able to optimize AGV movement in terms of guide and strategy?; and 3) is the use of artificial intelligence (AI) in AGV navigation techniques able to increase system performance? As a result, numerous techniques for AGV navigation have been developed globally. On the other hand, the most popular navigation approaches are provided below for more research.

Aiming at the problems of poor flexibility, complicated path maintenance and poor positioning performance in the current guidance technology of AGV, this paper designs and implements a new navigation system of automated guided vehicle (AGV) based on AprilTags2 auxiliary positioning. The system uses the Robot Operating System (ROS) as a platform to develop AGV navigation functions. The navigation system comprises two parts: the hardware and software layers. Firstly, hardware selection is performed after considering the actual requirements, performance, cost, and other factors in the hardware layer. Simultaneously, the AGV chassis and single-steering wheel walking mechanism are built to provide a stable and flexible operating platform for the software layer. Secondly, the software layer design includes two parts, namely the ROS navigation planning end and AprilTags2 detection. The ROS planning navigation end performs the design of four functional modules (i.e., AGV map construction, autonomous positioning, path planning, and path tracking), while the AprilTags2 detection part obtains the visual positioning pose of AGV by setting the AprilTags2 on each site and using the Kinect1 camera to detect. A more accurate AGV pose can be obtained by merging the former pose with the kinematic estimated pose. Finally, the two groups of positioning errors of the system before and after adding the AprilTags2 auxiliary positioning are tested and compared. The experimental results show that the navigation accuracy of the proposed system with AprilTags2 auxiliary positioning is significantly improved.

This paper presents the design and development of a mecanum wheel-based Automated Guided Vehicle (AGV), serving as a test platform for autonomous system research. The study highlights the essential elements of an Automated Guided Vehicle (AGV) and provides a thorough review of the literature. It explores the information, electrical, and mechanical actuator systems critical to the AGV's functioning. The AGV's navigation system is powered by a Jetson Nano minicomputer, with high-level software development implemented using the Robot Operating System (ROS). The software design focuses on modularity because it facilitates easier changes and maintenance, which is essential for research and rapid testing. The DC motors control subsystem controls AGV mobility by adjusting wheel speed and direction. These motors have incremental encoders that indicate wheel rotation. The Jetson Nano is used in the framework to process data from various sensors and monitor the AGV's performance. Experimental validation was carried out in both static and dynamic environments, where the AGV successfully executed autonomous navigation tasks with an average localization error, motion time, and path deviation. These results demonstrate the system's reliability, accuracy, and potential as a flexible test platform for developing and benchmarking autonomous navigation algorithms in research and educational contexts.

Automated Guided Vehicles (AGV) with magnetic navigation and electromagnetic navigation have lower path flexibility due to the need to lay the path ahead of time. The Simultaneous Localization and Mapping (SLAM) is mature in technology theory, so that this technique is applied to the AGV navigation. The laser sensor is used to acquire the information in the location environment for immediate location and mapping, so that carry out the path planning and complete the AGV autonomous navigation. Extended Kalman Filter (EKF) is the most common solution for laser SLAM. Firstly, this paper analyzes the research results of EKF-SLAM at this stage and its limitations. Secondly, it discusses the application of EKF-SLAM in AGV and the development direction in the next stage. By comparative analysis shows that laser EKF-SLAM can meet the requirements of AGV navigation. The development of EKF-SLAM algorithm in positioning, mapping accuracy and the ability to solve interferences in dynamic environment has an important reference for the reliability of AGV navigation.

- Automated Guided Vehicles (AGVs) play a vital role in modern industries for efficient material handling and internal logistics. Conventional AGV systems rely on fixed infrastructure such as embedded wires, magnetic tapes, or predefined tracks, which limits flexibility and increases installation cost. This paper presents the design and implementation of an Intelligent Automated Guided Vehicle (IAGV) using machine vision and image-based visual serving. The proposed system utilizes image processing techniques to recognize directional signs and navigate autonomously in a dynamic shop-floor environment. The mechanical structure is designed using a three-wheel tadpole configuration to enhance maneuverability and reduce rolling resistance. Structural analysis using finite element analysis (FEA) validates the load-carrying capacity of the vehicle. Experimental results demonstrate that the proposed IAGV provides improved flexibility, safety, and adaptability compared to conventional AGV systems, making it suitable for modern industrial environments. Key Words: Automated Guided Vehicle (AGV), Machine Vision, Image Processing, Visual Servoing, Intelligent Systems.

This paper considers the problem of real-time dispatching of a fleet of automated guided vehicles (AGVs) with battery constraints. AGVs must be immediately assigned to transport requests, which arrive randomly. In addition, the AGVs must be repositioned and recharged, awaiting future transport requests. Each transport request has a soft time window with late delivery incurring a tardiness cost. This research aims to minimize the total costs, consisting of tardiness costs of transport requests and travel costs of AGVs. We extend the existing literature by making a distinction between parking and charging nodes, where AGVs wait idle for incoming transporting requests and satisfy their charging needs, respectively. Also, we formulate this online decision-making problem as a Markov decision process and propose a solution approach based on deep reinforcement learning. To assess the quality of the proposed approach, we compare it with the optimal solution of a mixed-integer linear programming model that assumes full knowledge of transport requests in hindsight and hence serves as a lower-bound on the costs. We also compare our solution with a heuristic policy used in practice. We assess the performance of the proposed solutions in an industry case study using real-world data.

With the booming development of logistics, manufacturing and warehousing fields, the autonomous navigation and intelligent obstacle avoidance technology of automated guided vehicles (AGVs) has become the focus of scientific research. In this paper, an enhanced deep reinforcement learning (DRL) framework is proposed, aiming to empower AGVs with the ability of autonomous navigation and obstacle avoidance in the unknown and variable complex environment. To address the problems of time‐consuming training and limited generalisation ability of traditional DRL, we refine the twin delayed deep deterministic policy gradient algorithm by integrating adaptive noise attenuation and dynamic delayed updating, optimising both training efficiency and model robustness. In order to further strengthen the AGV's ability to perceive and respond to changes of a dynamic environment, we introduce a distance‐based obstacle penalty term in the designed composite reward function, which ensures that the AGV is capable of predicting and avoiding obstacles effectively in dynamic scenarios. Experiments indicate that the AGV model trained by this algorithm presents excellent autonomous navigation capability in both static and dynamic environments, with a high task completion rate, stable and reliable operation, which fully proves the high efficiency and robustness of this method and its practical value.

With the development of logistics industry, the logistics automated warehouse is ahead in the race toward the direction of multi-user, multi-variety and flexibility automated warehouse. Traditional AGVs (automatic guided vehicle) are difficult to meet the new demands of logistics warehouse. This paper presents a method to extract the deviation of AGV's navigation path based on threshold segmentation of chromatic aberration according to the actual needs. The paper realized high precision path tracking of robot by the path tracking controller adopting the idea of beforehand closed loop control. Finally, the logistics warehouse environment was simulated, and the navigation performance of the designed AGV was fully tested. The experimental results show that the visual navigation AGV features of flexible navigation, high precision and smooth operation and can satisfy the demand of flexible storage.

Automated guided vehicles (AGVs) are widely used for transportation in flexible job shop (FJS) systems, and their transportation task scheduling has the same substantial impact on production efficiency as machine scheduling does. However, traditional FJS scheduling methods often prioritize job sequencing and machine selection while ignoring the impact of AGV transportation, resulting in suboptimal scheduling solutions and even difficulties in implementation. To address this issue, this paper formulates a cooperative scheduling model by introducing the AGV scheduling problem into the classical FJS scheduling problem, abbreviated as the FJS-AGV problem, with the objective of minimizing the makespan. With respect to the FJS-AGV problem, a heuristic-assisted deep Q-network (HA-DQN) algorithm is proposed, which leverages heuristic rules to enable the decision agent to perform multiple actions at each decision point, which includes determining the responses to the following questions: Which operation should be processed next? On which machine? By which AGV? This decision mechanism enables the agent to make more informed decisions, leading to improved performance and resource allocation in the FJS-AGV system. The practicability of the proposed FJS-AGV model and the efficiency of the HA-DQN algorithm in solving the FJS-AGV problem are verified through various international benchmarks. Specifically, when solving instances in a large benchmark, the HA-DQN algorithm yields a significant 12.63% reduction in makespan compared with that when traditional heuristics are employed.

Automated guided vehicles and autonomous mobile robots for recognition and tracking in civil engineering

As the need for efficient warehouse logistics has increased in manufacturing systems, the use of automated guided vehicles (AGVs) has also increased to reduce travel time. The AGVs are controlled by a system using laser sensors or floor-embedded wires to transport pallets and their loads. Because such control systems have only predefined palletizing strategies, AGVs may fail to engage incorrectly positioned pallets. In this study, we consider a vision sensor-based method to address this shortcoming by recognizing a pallet’s position. We propose a multi-task deep learning architecture that simultaneously predicts distances and rotation based on images obtained from a visionary sensor. These predictions complement each other in learning, allowing a multi-task model to learn and execute tasks impossible with single-task models. The proposed model can accurately predict the rotation and displacement of the pallets to derive information necessary for the control system. This information can be used to optimize a palletizing strategy. The superiority of the proposed model was verified by an experiment on images of stored pallets that were collected from a visionary sensor attached to an AGV.

A motion planning method that integrates global planning and local control is proposed in this paper, aiming to enable Automated Guided Vehicle (AGV) to navigate automatically in two-dimensional environment. The layered cost maps are utilized and divided into global cost map and local cost map to respectively provide environment information for the global planning and local control. Specifically, the global planning is based on the path planning algorithm to establish a global optimal path on the global cost map. Under the guidance of global path, local control generates an optimal collision free trajectory with temporal information on the local cost map, which can avoid contact with surrounding obstacles. The local control operates in real-time and consider temporal information, thereby it can directly generate commands for the underlying robot motion controller. Both the simulation and experimental results are presented in support of the feasibility and effectiveness of the proposed motion planning.

Automated guided vehicle dispatching and routing integration via digital twin with deep reinforcement learning

In the context of the continuous development and maturity of intelligent manufacturing and intelligent logistics, it has been observed that the majority of vehicle maintenance in EMU trains still relies on traditional methods, which are characterized by excessive manual intervention and low efficiency. To address these deficiencies, the present study proposes the integration of Automatic Guided Vehicles (AGVs) to improve the traditional maintenance processes, thereby enhancing the efficiency and quality of vehicle maintenance. Specifically, this research focuses on the scenario of the maintenance workshop in EMU trains and investigates the task allocation problem for multiple AGVs. Taking into consideration factors such as the maximum load capacity of AGVs, remaining battery power, and task execution time, a mathematical model is formulated with the objective of minimizing the total distance and time required to complete all tasks. A multi-population genetic algorithm is designed to solve the model. The effectiveness of the proposed model and algorithm is validated through simulation experiments, considering both small-scale and large-scale scenarios. The results indicate that the multi-population genetic algorithm outperforms the particle swarm algorithm and the genetic algorithm in terms of stability, optimization performance, and convergence. This research provides scientific guidance and practical insights for enterprises adopting task allocation strategies using multiple AGVs.

This paper presents the development of an innovative indoor localization system that leverages a self-rotating inertial sensor for direction estimation and an LSTM-based deep learning model for distance estimation. The primary objective is to enhance the positioning accuracy of two-dimensional moving objects, such as Automatic Guided Vehicles (AGVs), in environments where GPS is unavailable. Traditional indoor localization methods, including wireless, Simultaneous Localization and Mapping (SLAM), and inertial navigation, each have limitations in terms of accuracy and robustness. The proposed system overcomes these limitations in inertial navigation by combining a unique hardware configuration with deep learning techniques. The system’s effectiveness was validated through various performance evaluations, demonstrating significant improvements in direction and distance estimation accuracy. Specifically, the direction error was reduced from 52° to 5° over a 30-minute evaluation period. Additionally, it was confirmed that the average error in estimating the travel distance for straight and rectangular paths was reduced from over 3% with conventional methods to less than 1%.