Path planning for active SLAM based on deep reinforcement learning under unknown environments

Right invariant extended Kalman filter (RIEKF) based simultaneous localization and mapping (SLAM) proposed recently has shown to be able to produce more consistent SLAM estimates as compared with traditional EKF based SLAM methods, including some improved EKF SLAM methods such as observability constrained-EKF (OC-EKF) SLAM. Latest results have demonstrated that its performance is very close to optimization based SLAM algorithms such as iSAM. In this paper, we propose to use RIEKF SLAM algorithm in active SLAM where both the predicted SLAM results for choosing control actions and the actual estimated SLAM results applying the selected control actions are computed using RIEKF algorithms. The advantages over traditional EKF based active SLAM are the more accurate and consistent predicted uncertainty estimates which result in robustness of the active SLAM algorithm. The advantages over optimization based active SLAM is the reduced computational cost. Simulation results are presented to validate the advantages of the proposed algorithm³.

In this paper, we present a novel integrated vehicular system using collaborative unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) for autonomous exploration, mapping, and navigation in GPS-denied 3-dimensional (3-D) unknown environments. The system implements a novel two-layered exploration strategy to decompose the perception task into a coarse exploration layer and a fine mapping layer. The coarse exploration makes use of a UGV to carry out fast autonomous exploration and active 2.5D simultaneous localization and mapping (SLAM) to generate a coarse environment model, which serves as a navigation reference for subsequent complementary 3-D fine mapping conducted by a UAV. The two layers share a novel optimized exploration path planning and navigation framework, which provides optimal exploration paths and integrates the collaborative exploration and mapping efforts through an OctoMap-based volumetric motion planning interface. The proposed system provides an efficient pipeline of fast environment perception taking advantages of the agility of the UAVs as well as the powerful computation resource aboard UGVs, also allowing assistive local perception with augmented object information when necessary. The effectiveness of our system is verified by both simulations and experiments, which demonstrate its capability of implementing heterogeneous UAV and UGV collaborative exploration and structural reconstruction of the environments through active SLAM, providing optimized perception for navigation tasks.

This paper proposes a novel active Simultaneous Localization and Mapping (SLAM) method with continuous trajectory optimization over a stochastic robot dynamics model. The problem is formalized as a stochastic optimal control over the continuous robot kinematic model to minimize a cost function that involves the covariance matrix of the landmark states. We tackle the problem by separately obtaining an open-loop control sequence subject to deterministic dynamics by iterative Covariance Regulation (iCR) and a closed-loop feedback control under stochastic robot and covariance dynamics by Linear Quadratic Regulator (LQR). The proposed optimization method captures the coupling between localization and mapping in predicting uncertainty evolution and synthesizes highly informative sensing trajectories. We demonstrate its performance in active landmark-based SLAM using relative-position measurements with a limited field of view.

Simultaneous Localization and Mapping (SLAM)consists in the concurrent construction of a model of the environment (the map), and the estimation of the state of the robot moving within it. The SLAM community has made astonishing progress over the last 30 years, enabling large-scale real-world applications, and witnessing a steady transition of this technology to industry. We survey the current state of SLAM. We start by presenting what is now the de-facto standard formulation for SLAM. We then review related work, covering a broad set of topics including robustness and scalability in long-term mapping, metric and semantic representations for mapping, theoretical performance guarantees, active SLAM and exploration, and other new frontiers. This paper simultaneously serves as a position paper and tutorial to those who are users of SLAM. By looking at the published research with a critical eye, we delineate open challenges and new research issues, that still deserve careful scientific investigation. The paper also contains the authors' take on two questions that often animate discussions during robotics conferences: Do robots need SLAM? and Is SLAM solved?

Simultaneous localization and mapping (SLAM) consists in the concurrent construction of a model of the environment (the map), and the estimation of the state of the robot moving within it. The SLAM community has made astonishing progress over the last 30 years, enabling large-scale real-world applications and witnessing a steady transition of this technology to industry. We survey the current state of SLAM and consider future directions. We start by presenting what is now the de-facto standard formulation for SLAM. We then review related work, covering a broad set of topics including robustness and scalability in long-term mapping, metric and semantic representations for mapping, theoretical performance guarantees, active SLAM and exploration, and other new frontiers. This paper simultaneously serves as a position paper and tutorial to those who are users of SLAM. By looking at the published research with a critical eye, we delineate open challenges and new research issues, that still deserve careful scientific investigation. The paper also contains the authors' take on two questions that often animate discussions during robotics conferences: Do robots need SLAM? and Is SLAM solved?

Path planning refers to that the mobile robot can obtain the surrounding environment information and its own state information through the sensor carried by itself, which can avoid obstacles and move towards the target point. Deep reinforcement learning consists of two parts: reinforcement learning and deep learning, mainly used to deal with perception and decision-making problems, has become an important research branch in the field of artificial intelligence. This paper first introduces the basic knowledge of deep learning and reinforcement learning. Then, the research status of deep reinforcement learning algorithm based on value function and strategy gradient in path planning is described, and the application research of deep reinforcement learning in computer game, video game and autonomous navigation is described. Finally, I made a brief summary and outlook on the algorithms and applications of deep reinforcement learning.

This paper reports on an active simultaneous localization and mapping (SLAM) framework that leverages the Bayes tree data structure for efficient planning. Evaluating information-theoretic objective functions in the context of active SLAM is a very expensive process that requires significant computational overhead. The contributions of this work involve exploiting the structure of the planning problem integrated with SLAM via the Bayes tree graphical model. Specifically, we propose a constrained variable ordering and subtree caching scheme that reduce computational complexity by eliminating redundant computations between candidate actions that are similar. We also propose an active SLAM framework that utilizes these concepts, and demonstrate the benefits of the approach with an underwater robot performing visual SLAM in a hybrid simulation environment.

As mobile robots are becoming more and more widely used, it is of great significance to design an efficient path planning method for multi-robot systems (MRS) that can adapt to complex and unknown environments. In this paper, we present a deep reinforcement learning (DRL) based method to navigate the MRS with collision avoidance in unknown dynamic environments, which consists of centralized learning and decentralized executing paradigm. The proposed policy maps original laser information into robot control commands without constructing global maps. The learned policy is tested in Gazebo environments with three robot systems, which shows the effective performance in terms of success rate, extra time rate, and formation maintenance rate.

Autonomous service robots are gaining popularity as they find application in different sectors such as logistics, hospitality, medical, space exploration and military. It is widely seen that autonomous service robots help increase efficiency, reduce cost and ensure safer operations. Automating material handling operations in highly dynamic indoor environments is risky and involves use of complex algorithms. However, continuously advancing technologies and machine programming tend to overcome these challenges. The three essential functions of an autonomous robot are to identify its instantaneous location in three-dimensional space, generate the environment map and decide an optimal path for locomotion. Simultaneous Localization and Mapping (SLAM) is a widely used algorithm for navigation in unfamiliar environments. This chapter presents a cost-effective autonomous ground robot for automating material handling, transportation and other repetitive tasks in dynamically changing surroundings. The robot is programmed to map the indoor environment and localize itself in the workspace with the help of SLAM and Adaptive Monte Carlo Localization (AMCL). In addition, the robot can manoeuvre through the indoor workspace by avoiding static and dynamic obstacles. This work employs the GMapping open-source package in the Robot Operating System (ROS) for performing SLAM with the aid of a laser range finder, wheel odometry and inertial data. The GMapping algorithm uses a probabilistic approach to construct a map and estimate the robot’s location in two-dimensional space. In addition, this chapter presents an autonomous navigation feature with the help of the AMCL method for localizing the robot within an uncertain environment and making the robot move to its goal using move base actions in ROS. The path planning of the mobile robot is classified into global and local plans. The global plan is responsible for planning the path from the start point to the endpoint. The local plan is responsible for planning the actual movement of the robot by avoiding any dynamic obstacles not mapped earlier. This work uses the Dijistra algorithm for the global and the dynamic window approach (DWA) algorithm for the local path planning. The optimization algorithm minimizes the error between the global and local paths. The simulation study of the proposed algorithm is performed in a virtual environment called Gazebo in ROS, which is nearly similar to a real-time indoor workspace. Further, a robot prototype is fabricated, and its performance is evaluated in an unknown environment. The robot prototype is equipped with brushed DC motors with optical encoder feedback for wheel odometry (ODOM), Light Detection and Ranging (LiDAR), and Inertial Measurement Unit (IMU) sensors controlled using a Jetson Nano. The simulation and experimental results showcase the robot’s successful two-dimensional localization and navigation in the factory environment. The current system has the potential for indoor assistance during material transfer and logistic applications. Further upgrades to the system can make it usable for Automated Storage and Retrieval System (ASRS) applications in warehouses/factories.

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.

In the absence of GPS signals, Simultaneous Localization and Mapping (SLAM) technology enables unmanned systems to accomplish map construction and autonomous navigation in unknown environments. At present, there are mainly methods for SLAM based on sensors such as lidar, camera, ultrasonic, Inertial Measurement Unit (IMU) and odometer for location environment. However, there are certain limitations in using a single sensor for SLAM in unknown environment. In order to solve the above problems, this paper summarizes the key problems faced by mainstream SLAM at present, and proposes a SLAM method based on multi-sensor fusion, which can make full use of the advantages of various sensors from the hardware structure and make up for the shortcomings of using a single sensor. Finally, the proposed method was verified on the platform of “JAC Electric Vehicle”, and the experimental results show that the proposed method can effectively avoid the influence caused by signal weakening, improve the positioning accuracy, and the system has stronger robustness and better tracking performance.

Simultaneous Localization and Mapping (SLAM) methods are used in autonomous systems to determine their locations in unknown environments and map these environments. Autonomous systems need to act autonomously without external intervention. These methods are widely used in robotics and AR/VR applications. Gaussian Splatting SLAM is a Visual SLAM method that performs mapping and localization using depth and RGB images and uses Gaussian structures for scene representation. Popular datasets such as TUM-RGBD, Replica, and Scannet++ are used in the performance evaluation and testing of the visual SLAM methods. However, the depth images in the TUM-RGBD dataset are of lower quality than other datasets. This problem negatively affects the depth data's accuracy and reduces the quality of mapping results. In this study, to increase the quality of depth images, the features of depth images were corrected using the median filter, which is the depth smoothing method, and a cleaner depth dataset was obtained. The new dataset obtained was processed using the Gaussian Splatting SLAM method, and better metric results (PSNR, SSIM, and LPIPS) were obtained compared to the original dataset. As a result, in the dataset with corrected features, an improvement of 8.08% in the first scene and 4.69% in the second scene was achieved according to metric values compared to the original dataset.

Uncertainty remains a central challenge in robotic navigation, exploration, and coordination. This paper examines how Partially Observable Markov Decision Processes (POMDPs) and their decentralized variants (Dec-POMDPs) provide a rigorous foundation for decision-making under partial observability across tasks such as Active Simultaneous Localization and Mapping (A-SLAM), adaptive informative path planning, and multi-robot coordination. We review recent advances that integrate deep reinforcement learning (DRL) with POMDP formulations, highlighting improvements in scalability and adaptability as well as unresolved challenges of robustness, interpretability, and sim-to-real transfer. To complement learning-driven methods, we discuss emerging strategies that embed probabilistic reasoning directly into navigation, including belief-space planning, distributionally robust control formulations, and probabilistic graph models such as enhanced probabilistic roadmaps (PRMs) and Canadian Traveler Problem-based roadmaps. These approaches collectively demonstrate that uncertainty can be managed more effectively by coupling structured inference with data-driven adaptation. The survey concludes by outlining future research directions, emphasizing hybrid learning–planning architectures, neuro-symbolic reasoning, and socially aware navigation frameworks as critical steps toward resilient, transparent, and human-centered autonomy.

Unmanned vehicles in urban structured environments have become the main battlefields of related companies all over the world. By relying on high-precision maps established in advance, unmanned vehicles can perform well enough. However, this method is difficult to apply to unknown off-road environment which cannot obtain high-precision maps in advance, such as transportation and target search of dangerous scenes in the wild. In this paper, we introduce a novel and efficient real-time path planning method for autonomous driving in unknown environments with sparse waypoints. The contributions of this paper are as follows. First, we present a goal point extraction algorithm to extract the goal configuration in free space for local path planning. It is extremely suitable for the sparse waypoints situation, in which local perception map may not contain any waypoints. Second, we present a novel path re-planning algorithm for large-scale environment without SLAM(Simultaneous Localization And Mapping) by building a topology map constantly. Our method not only overcame the failure of path planning due to the limited on-board sensing range, but also avoided high computing resource consumption brought by the large-scale construction maintenance of SLAM. A large number of challenging real off-road scenarios show that the proposed method can effectively solve the problem that unmanned vehicles cannot plan to travel due to sparse waypoints, and give real-time effective planning results when unmanned vehicles need to carry out large-scale path re-planning.

The active SLAM problem considered in this paper aims to plan a robot trajectory for simultaneous localization and mapping (SLAM) as well as for an area coverage task with robot pose uncertainty. Based on a model predictive control (MPC) framework, these two problems are solved respectively by different methods. For the uncertainty minimization MPC problem, based on the graphical structure of the 2D feature-based SLAM, a non-convex constrained least-squares problem is presented to approximate the original problem. Then, using variable substitutions, it is further transformed into a convex problem, and then solved by a convex optimization method. For the coverage task considering robot pose uncertainty, it is formulated and solved by the MPC framework and the sequential quadratic programming (SQP) method. In the whole process, considering the computation complexity, we use linear SLAM, which is a submap joining approach, to reduce the time for planning and estimation. Finally, various simulations are presented to validate the effectiveness of the proposed approach.