Search

This comprehensive review focuses on the Autonomous Driving System (ADS), which aims to reduce human errors that are the reason for about 95% of car accidents. The ADS consists of six stages: sensors, perception, localization, assessment, path planning, and control. We explain the main state-of-the-art techniques used in each stage, analyzing 275 papers, with 162 specifically on path planning due to its complexity, NP-hard optimization nature, and pivotal role in ADS. This paper categorizes path planning techniques into three primary groups: traditional (graph-based, sampling-based, gradient-based, optimization-based, interpolation curve algorithms), machine and deep learning, and meta-heuristic optimization, detailing their advantages and drawbacks. Findings show that meta-heuristic optimization methods, representing 23% of our study, are preferred for being general problem solvers capable of handling complex problems. In addition, they have faster convergence and reduced risk of local minima. Machine and deep learning techniques, accounting for 25%, are favored for their learning capabilities and fast responses to known scenarios. The trend towards hybrid algorithms (27%) combines various methods, merging each algorithm's benefits and overcoming the other's drawbacks. Moreover, adaptive parameter tuning is crucial to enhance efficiency, applicability, and balancing the search capability. This review sheds light on the future of path planning in autonomous driving systems, helping to tackle current challenges and unlock the full capabilities of autonomous vehicles. [ABSTRACT FROM AUTHOR]

In several applications, a robot moving from a start to a goal location is required to gather data along its path (e.g., a video feed in a monitoring scenario). The robot can have at its disposal only a limited amount of memory to store the collected data, in order to contain costs or to avoid that sensitive data fall into the hands of an attacker. This poses the need of periodically delivering the data to a Base Station (BS) through a deployed communication infrastructure that, in general, is not available everywhere. In this paper, we study this scenario by considering a variant of the shortest path problem (which we prove to be NP-hard) where the robot acquires information along its path, stores it into a limited memory buffer, and ensures that no information is lost by periodically communicating data to the BS. We present and evaluate an optimal exponential-time algorithm, an efficient feasibility test, and a polynomial-time heuristic algorithm. • A robot gathers data along its path with limited emory and restricted communication. • We study this scenario as a constrained variant of the shortest path problem. • We show that the problem is NP-hard and propose optimal and heuristic algorithms. [ABSTRACT FROM AUTHOR]

Intelligent path planning is a significant tool for field of industrial robot. This field has attracted the attention of numerous researchers due to the great market demands, broad application prospects, and large potential development. Due to the limitation of neighborhood, the path search by the original A* algorithm is more likely to fail, and the solved path may contain too many local paths. In this study, an improved A* algorithm is proposed to solve the robot path planning problem. The first improvement of the advanced method is the local path between the current node and the goal node, which is planned before the next search in the neighborhood of the current node. And the local path will be adopted directly if it is safe and collisionless. The second advantage of this method is the utilization of post-processing stage to optimize the resulting path, by straightening the local path to reduce the number of local paths as well as the path length. In order to verify the theoretical advantages of the improved A* algorithm, a series of two-dimensional figures of the robot task was presented in this paper. In addition, some comparative experiments in the virtual and real robot manipulator platform are performed to examine the improved A* algorithm. Experimental results show that the search success rate of the improved A* algorithm is higher than the original A* algorithm, along with a shorter and smoother path could be obtained by the improved A* algorithm. Therefore, the success rate of robot path planning and the optimal extent of the robot path are effectively improved by the improved A* algorithm. [ABSTRACT FROM AUTHOR]

This paper focuses on the mobile robot path planning problem of optimizing the performance metrics of bidirectional A* algorithm in randomized two-dimensional map environments. An algorithm called direction constraints adaptive extended bidirectional A* (DCAE-BA*), which is an improvement of the traditional target dynamic bidirectional A* algorithm (TTD-BA*), is proposed to improve the performance metrics of the algorithm. Regarding the improvement, we propose the adaptive extension method and the direction-constrained optimal node extension method (DCONE). Simulation experiments were conducted for DCAE-BA*, TTD-BA* and traditional A* algorithm (A*) in a large number of random two-dimensional map environments. The simulation experimental scenarios consider four types of start and end point relative directions and three obstacle proportions to objectively and comprehensively evaluate the performance of the proposed algorithms. The results show that different scenarios have a significant impact on the algorithm performance metrics. Finally, the overall performance of the proposed algorithm is evaluated with a large number of experiments in "random" scenarios, and the results show that DCAE-BA* obtains significantly better search time for all three obstacle proportions, and better path length and number of expanded nodes for 10% and 25% obstacle proportions. The effectiveness of the proposed DCAE-BA* algorithm is demonstrated, which provides an essential reference for the path planning of mobile robots in a random 2D map environment. • DCAE-BA* is improved by adaptive extension and direction-constrained optimal node extension (DCONE). • The Start-End Dir and obstacle proportions have a significant impact on the algorithm metrics. • Overall, the DCAE-BA* has the highest computational efficiency for all three types of obstacle proportion. • DCAE-BA* optimizes the path length in the "diagonal" scenario most significantly. • As the proportion of obstacles increases, the metric optimization effect decreases. [ABSTRACT FROM AUTHOR]

This paper presents a mission system and the therein implemented algorithms for path planning in a time-varying environment based on graph methods. The basic task of the introduced path planning algorithms is to find a time-optimal path from a defined start position to a goal position with consideration of the time-varying ocean current for an autonomous underwater vehicle (AUV). Building on this, additional practice-oriented considerations in planning are discussed in this paper. Such points are the discussion of possible methods to accelerate the algorithms and the determination of the optimal departure time. The solutions and algorithms presented in this paper are focused on path planning requirements for the AUV “SLOCUM Glider”. These algorithms are equally applicable to other AUVs or aerial mobile autonomous systems. [ABSTRACT FROM AUTHOR]

To date, a large number of optimization algorithms have been presented for Autonomous Underwater Vehicle (AUV) path planning. However, little effort has been devoted to compare these techniques. In this paper, an quantum-behaved particle swarm optimization (QPSO) algorithm is introduced for solving the optimal path planning problem of an AUV operating in environments with ocean currents. An extensive study of the most important optimization techniques applied to optimize the trajectory for an AUV in several test scenarios is presented. Extensive Monte Carlo trials were also run to analyse the performance of these optimization techniques based on solution quality and stability. The weaknesses and strengths of each technique have been stated and the most appropriate algorithm for AUV path planning has been determined. [ABSTRACT FROM AUTHOR]

This paper addresses the path planning problem for robotic applications using homotopy classes. These classes provide a topological description of how paths avoid obstacles, which is an added value to the path planning problem. Homotopy classes are generated and sorted according to a lower bound heuristic estimator using a method we developed. Then, the classes are used to constrain and guide path planning algorithms. Three different path planners are presented and compared: a graph-search algorithm called Homotopic A ∗ (HA ∗ ), a probabilistic sample-based algorithm called Homotopic RRT (HRRT), and a bug-based algorithm called Homotopic Bug (HBug). Our method has been tested in simulation and in an underwater bathymetric map to compute the trajectory of an Autonomous Underwater Vehicle (AUV). A comparison with well-known path planning algorithms has also been included. Results show that our homotopic path planners improve the quality of the solutions of their respective non-homotopic versions with similar computation time while keeping the topological constraints. [ABSTRACT FROM AUTHOR]

Most of the existing path planning algorithms require a prior global map. Although there have been some algorithms proposed for unknown environments, they can only deal with those which just have several hidden obstacles in a roughly known global map. In order to improve the efficiency of robot's path planning without prior global maps, this paper proposes a Convolutionally Evaluated Gradient First Search (CE-GFS) path planning algorithm. It allows the robot to collect environmental information and complete path planning simultaneously. Firstly, the Gradient First Search (GFS) algorithm is proposed based on the gradient score parameter, with which the conventional cost function is replaced. The GFS can adapt to any moving direction through the environmental information surrounding the mobile robot and computing the gradient score parameter. Secondly, CE-GFS path planning algorithm is proposed based on GFS and convolutional evaluation method. The CE-GFS helps the robots to evaluate the efficiency of the global path and get global perception ability, so that they are prevented from going astray, which can significantly improve the efficiency of path planning. Finally, several simulation tests and field tests have been carried out. The test results show that convolutional evaluation improves the efficiency of CE-GFS by 86.94% on average compared with GFS at the price of 0.18% decrease in path planning success rate. Moreover, the time cost of the proposed CE-GFS algorithm is 42.18% less than that of FAR-Planner in some special cases. • Our CE-GFS gets rid of this dependence on an exploration algorithm, and allows the robot to combine the exploration and path planning stages. • Cost functions in conventional grid map algorithms are replaced with gradient score in our algorithm, to make the robot more flexible. • Convolutional evaluation method is proposed to evaluate the efficiency of the global path through local information, which can prevent the robot from going astray. [ABSTRACT FROM AUTHOR]

This paper presents a collision avoidance algorithm for stabilized gun turrets and its real-time implementation. With the help of new collision avoidance algorithm, all types of turrets can be driven more efficiently and safely according to the specified speed, acceleration and jerk limits. Even in situations such as avoiding obstacles, deceleration/acceleration, if the user issues new commands which does not cause a collision, the algorithm starts to apply the new commands providing flexibility to the user. Since all possible worst scenarios are examined one by one, it is guaranteed that the algorithm provides collision free motion in both simulations and real-time tests. A configuration space where worst scenarios can occur is created for the performance measurement of the algorithm, and the same space is used in all tests. By giving different speed commands in the specified configuration space, the performance of the algorithm at different speeds is observed on the stabilized gun turret. For the measurement of the performance under the noisy speed commands, a custom noisy speed command of about 1000 s is created and both simulation and real-time tests are performed. As a result of these tests, it is shown that there is no collision. Finally, by adding cascade position control loop, the departure from the starting point to the desired target point is achieved without any collision. The most important feature that distinguishes this algorithm from others is both speed and position can be controlled and during transition phase, the target point can be changed instantly. In addition, no target position is required for the system to move collision-free, only axis speed commands are sufficient. Since the algorithm does not intervene in the speed and torque loops in contrast to potential field-based methods, it can be added to ready-to-use systems by manipulating only the speed references. • A novel collision avoidance algorithm for gun turrets is created. • No need to specifying any target position to move the system without collision. • The system acts in compliance with all speed, acceleration and jerk limits. • By adding cascade position loop, it can be used as global path planning algorithm. • The algorithm does not intervene in the speed and torque. [ABSTRACT FROM AUTHOR]

The paper presents a novel path planning methodology based on the directional graph search and the geometry curve for the Automated Valet Parking (AVP) system. The whole path planning methodology is divided into three parts including the global path planning, the path coordination strategy and the parking path planning. Firstly, the global path planning is triggered to find a path from the parking slot entrance to the rough location of the assigned parking spot. A novel directional Hybrid A* algorithm is proposed to generate the global path efficiently without redundant searches, such as the dead end. Afterwards, the path coordination strategy gives a transitional path to connect the end node of the global path to the parking planning start node. The transitional path is composed of geometry curves including arcs and line segments based on the optimal parking start node. Finally, the parking path planning generates a parking path to guide the vehicle from parking start node to the parking space. A modified C-type vertical parking path planning algorithm is utilized to generate the parking path, offering flexibility for choosing the parking start node. Simulation results based on Matlab and PreScan show that it takes less time for the proposed path planning algorithm to generate a feasible path for the AVP system compared with the general planning algorithm. The novel AVP path planning algorithm also has the potential for practical use. • A novel path planning methodology for Automated Valet Parking is proposed. • The global path and the parking path are merged smoothly. • The global path planning can generate path without redundant search. • The optimal parking start node corresponding to any feasible heading angle is calculated. • The proposed methodology shows stability and improvement in terms of computational time than others. [ABSTRACT FROM AUTHOR]

This paper is concentrated on path planning for robots working in a dynamic environment to satisfy real-time needs. An efficient bias-goal factor RRT (EBG-RRT), which is multiple-query sampling-based replanning algorithm, is proposed with rapid response and high success rate. Specifically, a relay node method is proposed to get a position where the robot and dynamic obstacles will be no-collision and help robots to move without suspended. Based on the relay node method, Connection strategy performs minimal modifications to maintain the interrupted path. In order to overcome the short of Waypoint Cache method, an efficient and optimal Waypoint Cache (EOWC) method is proposed to make use of potential cache information and find an optimal path to repair. The EOWC method is combined with the BG-RRT algorithm according to the iterative characteristics. Finally, the EBG-RRT algorithm is verified on ROS with Aubo-i5 manipulator. Simulation results provide the EBG-RRT algorithm is outperformed both in static and dynamic environments. • We describe a multiple-query sampling-based replanning algorithm. • The first study to introduce relay node method to help dynamic obstacle avoidance. • An efficient and optimal Waypoint Cache (EOWC) method is proposed to make use of potential cache information. • Rapid convergence to path solution both in static and dynamic environments. [ABSTRACT FROM AUTHOR]