Adaptive path planning and tracking considering road safety constraints based on model predictive control for autonomous vehicles

Navigation and path planning are among the central problems in the development of mobile and autonomous robots. Research in this field has been conducted for decades, and several methodologies have been proposed to solve these problems. In the field, these approaches are divided into classical or deterministic and non-deterministic or heuristic methods. The article provides a brief overview of typical representatives of both classes, as well as an extended review of methods based on artificial potential fields. Important characteristics of obstacle detection and avoidance algorithms include convergence, computation time, and memory requirements in the system. The need for convergence arises from the requirement to achieve a stable or desired state of the system. This time varies depending on the chosen algorithm, the nature of the task, and the initial conditions. The main goal is to reduce convergence time, i.e., to reach the desired state as quickly as possible. Computation time and memory requirements are important because the robot must respond to the working environment and changes in it in real-time, and autonomous robots usually have quite limited hardware resources. Therefore, these are also important characteristics when selecting a method for a specific task and robot. The modification of the classical artificial potential field method using the Gaussian function to describe repulsive forces is an example of optimizing the method for systems with constrained resources. As of the writing of the article, unmanned aerial vehicles with limited resources are beginning to be widely used, making such optimizations practically valuable. Among the considered methods, heuristic ones are relatively new and are increasingly finding practical application. Research at the time of writing focuses on optimizing existing algorithms and hybridization to improve efficiency. An example of such hybridization is the artificial potential field method using fuzzy logic. This combines the classical artificial potential field method with a heuristic approach—fuzzy logic. This leads to some complexity in the method but solves typical problems of the classical algorithm, such as local minima, and increases the optimality and smoothness of the path. Most of obstacle detection and avoidance algorithms are working with only one type of sensor, such as ultrasonic distance sensors, LIDAR, or cameras. Each sensor technology and corresponding algorithms have their advantages and disadvantages. A promising approach is to use several types of sensors and algorithms, combining the results of different algorithms to achieve a more optimal final result, so called sensor fusion. However, it should be noted that this approach will require more sophisticated hardware. As robots increasingly become part of everyday life, it is quite possible that they will start working in collaboratively and interacting to solve assigned tasks. The development of collaborative methods for obstacle avoidance and interaction between robots in a single working environment is also a promising research direction. In summary, the gradual robotization of many processes in everyday life or production generates a high demand for research in the field of mobile robotics in general and methods for obstacle detection, avoidance and path planning in particular. Key words: robotics, obstacle avoidance, path planning, artificial potential field, autonomous robots, mobile robots.

This paper introduces a control system intended for autonomous vehicles, which consists mainly of a trajectory re-planning module capable of obstacle avoidance and a tracking control module. In the trajectory re-planning module, the Model Predictive Control (MPC) and Artificial Potential Field (APF) method are applied in combination. Generated by the APF, the attractive and repulsive fields improve the effectiveness of obstacle avoidance in conjunction with the MPC. Also, a new obstacle avoidance function is developed within this module. In the tracking control module, an approach that integrates MPC with Extended Kalman Filter (EKF) is adopted. Due to the high performance of the EKF in state estimation, the control accuracy of the MPC is enhanced, which enables precise tracking of the planned path. The proposed control system is validated by conducting joint simulations via Matlab-Simulink and Carsim. According to the experimental results, the proposed system demonstrates effective performance in both obstacle avoidance and path tracking.

To realize fast and efficient path planning and tracking for intelligent mobile robot in certain environments, improved artificial potential field (APF) method and new fuzzy controllers are proposed. In the path planning, basic idea of APF is inherited, and integrating with simulated annealing algorithm (SA), local minimum problem of APF is solved. In the path tracking, according to the description of robot oriented positioning control, novel path tracking fuzzy controllers which based on human driving behavior have been designed. And a position- orientation alternate control method is presented. When position error is bigger, a position controller is used to achieve the fast tracking, or if the error is smaller, a posture controller is used. The two kinds of controllers take in turns according to the position error. Simulation results illustrate the effectiveness of the proposed methods.

Path planning and tracking of autonomous vehicles are generally two independent tasks accomplished through traffic environment modeling, reference trajectory generation and vehicle motion control. In this paper, we proposed a unified path planning and tracking method utilizing the optimization algorithm of the model predictive control to generate optimal reference trajectory and vehicle motion control concurrently. The vehicle’s surroundings including obstacle vehicles and road marks are firstly reconstructed based on the artificial potential field approach which generates a reference trajectory, then the total potential of the traffic environment is incorporated into the cost function of the model predictive controller. Therefore, the path planning and tracking of the vehicle can be unified for collision avoidance with moving or fixed obstacles in its surrounding traffic environment. The simulation shows that this unified path planning and tracking method is capable of accomplishing the obstacle avoidance for the vehicle during severe traffic scenarios.

Needle insertion is widely used in modern clinical practice, and flexible needles enable surgeons to adjust the needle insertion path to avoid obstacles and compensate for target drift. To increase the accuracy and safety of the needle insertion procedure, a path planning and tracking control method for flexible needle steering in dynamic environment is proposed in this paper. The path planning method is based on the artificial potential field algorithm and curve fitting algorithm, after several feasible paths with variant curvatures is obtained, they are furtherly optimized based on path length, obstacle clearance and control effort to select the best one. Simulations show that the proposed path planning method performs better than the current constant curvature method and the rapidly-exploring random tree (RRT) based method. By finely tuning the initial planned path, an intraoperative path replanning method is proposed, which are used to compensate for the drift of target and obstacles. Then to ensure the needle is inserted along the planned path, a tracking control method is adopted based on fuzzy logic algorithm. With the proposed intraoperative path planning and tracking control method, the robustness and accuracy of needle insertion in congested environment can be improved, and experiments also show that our proposed method achieves satisfactory insertion safety and targeting accuracy in dynamic environment. The obtained accuracy of 1.57 mm is sufficient for most clinical needle insertion applications.

A collaborative optimization scheme of obstacle avoidance and singularity avoidance path planning method is presented for redundant robot. An improved real-time minimum distance calculated method is presented and search the connect rod which easy to collision based on this minimum distance. Complete the obstacle avoidance based on the self-motion of the redundant manipulator on a null space and two obstacle avoidance parameters related to real-time minimum distance are introduced to improve optimization of obstacle avoidance. Adopt the DLS method to solve the problem that very high joint velocities in the vicinity of singular configuration. At last, through simulation of planar 3R redundant manipulator, the algorithm proves to be feasible and effective. Introduction The assembling work is more complicated as the miniaturization of 3C products and then the high speed manipulator was introduced by manufactures as auxiliary equipment of the assemble work to improve the assembly efficient and quality. The risk of man-machine collaboration is increased significantly because the collaborators and the robot working space overlap in large areas while assembling product in man-machine collaboration. Moreover, the usual inverse kinematics solutions based on Jacobian pseudo-inverse cause very high joint velocities in the vicinity of singular configurations, and it will lead to the manipulator deviate from expected trajectory and will have a certain influence on assembly quality. So if we can design a path planning method which can make the robot complete obstacle and singularity avoidance automatically, then it will make the man-machine collaboration system large-scale application in the field of 3C product assemble. Compared with the traditional manipulator the redundant manipulator has the additional degrees of freedom, makes it has the advantages of flexible operation and can complete obstacle and singularity avoidance. In the field of obstacle-avoiding of the manipulator, the main research method including the artificial potential field method,fuzzy method, neural network method, genetic algorithm, Probabilistic Roadmaps method,Rapidly-exploring Random Tree method.However, these methods have some shortcomings and the contradiction between the optimal path the planning time and the complexity of algorithm is very difficult to solve. In the area of singularity-avoiding of the manipulator, the main research methods such as the Damped least square method, SICQP method, normal forms method. But these method is difficult to solve the contradiction of the tracking accuracy and the complexity of algorithm. An improved calculation method is proposed in this paper to compute the minimum distance in order to improve the real-time performance of the system. Two obstacle avoidance parameters related to minimum distance are introduced and the self-motion in null space of the redundant manipulator is utilized to accomplish obstacle avoidance. The Damped Least Square method is adopted to optimize manipulator performance in the vicinity of singular configurations, complete the optimization between singularity avoidance and obstacle avoidance for redundant manipulator. At last through simulation of 3-DOF redundant manipulator, the scheme proves to be feasible and effective. 5th International Conference on Computer Sciences and Automation Engineering (ICCSAE 2015) © 2016. The authors Published by Atlantis Press 959 Traditional distance calculation method For the traditional minimum distance calculation method, the first step is calculating the minimum distance between obstacles and each manipulator connecting bar of the manipulator and then taking the minimum value. Take a 3-DOF, for example, as shown in Figure 1(a), the real-time minimum distance can be described as dmin=min{d1,d2,d3}. While the manipulator motion to some special position such as shown in Figure 1(b), the pedal of the obstacle to the manipulator connecting bar may fell on the extension of the bar, because the extension of the rod is not part of the manipulator, so it need to selected some mark point from the connecting rod and compute the distance between the mark point and the obstacle, and then taking the minimum value. But this minimum distance is not accurate, and this method will lead to huge calculation, low efficiency and longtime of computation.

Path planning is one of the most fundamental problems that must be solved before a robot can navigate and explore autonomously. Path planning needs to be integrated with path tracking to be applied to autonomous robots. This makes path tracking also important for autonomous robot navigation which cannot be separated from path planning. There are two path planning methods, the first is search-based method, the second is sampling-based method. Both have their own advantages, but the popular and commonly used sampling-based algorithm due to its fast convergence is preferred in path planning. The RRT* algorithm was developed. This improvement initiated a major civilization in sampling-based algorithms, namely parent node selection and rewiring in RRT. Although there has been an improvement in optimality, RRT* still doesn't provide the distance optimality value as expected, due to its character that is still adopted from RRT. The resulting path is still suboptimal and not smooth (jagged). On the other side, Path tracking has several methods, however, these path tracking methods are difficult to apply to autonomous robots and need to be adapted to the robot used. Based on the description above, there are still problems with path planning, namely paths that are still less than optimal and convergence that is still slow. This research will add a way to shorten the distance in the RRT* algorithm with the triangular inequality method. Meanwhile, for path tracking, we will apply the pose-to-pose method, which follows the waypoint that has been made by path planning.

In order to diminish the impacts of external disturbance such as parking speed fluctuation and model uncertainty existing in steering kinematics, this paper presents a parallel path tracking method for vehicle based on preview back propagation (BP) neural network PID controller. The forward BP neural network can adjust the parameters of PID controller in real time. The preview time is optimized by considering path curvature, change in curvature and road boundaries. A fuzzy controller considering barriers and different road conditions is built to select the starting position. In addition, a kind of path planning technology satisfying the requirement of obstacle avoidance is introduced. In order to solve the problem of discontinuous curvature, cubic B spline curve is used for curve fitting. The simulation results and real vehicle tests validate the effectiveness of the proposed path planning and tracking methods.

Addressing the conflict between path tracking and vehicle stability for autonomous driving vehicles, a new planning and control strategy has been proposed, which takes into account both vehicle dynamic stability and path tracking performance. At the local planning level, a model predictive control (MPC)‐based method for local path planning is adopted. Adaptive preview logic is utilized to dynamically adjust the prediction horizon size based on vehicle speed, preview trajectory point curvature, and side slip angle, in order to balance the path tracking performance and vehicle dynamic stability. To optimize the adaptive preview logic parameters, a particle swarm optimization (PSO) algorithm is employed for offline parameter optimization. Further, a two‐layer MPC path planning and tracking system was designed to verify this approach. Simulation experiments demonstrate that in complex scenarios such as lane changing and obstacle avoidance, the proposed strategy can effectively balance vehicle dynamic performance and path tracking accuracy.

With the development and widely use of unmanned aerial vehicles (UAVs) in recent years, the development of efficient path planning methods for automation has become crucial. Obstacle avoidance and path planning are the key components of UAV path planning. This article provides an overview of obstacle avoidance and path planning techniques for UAVs based on the artificial potential field method (APF method). This article begins with the explaining the principles of artificial potential field on this basis discusses its advantages and limitations. The article then summarizes the improvement strategies proposed by previous researchers to address issues like local minimum values and unreachable targets, such as introducing a new repulsive potential energy function, combining APF with other planning methods, and utilizing flow functions. Furthermore, it presents examples of the application and the performance of usage of these techniques in both static and dynamic environments. Based on this, the prospects and developing trend of UAV obstacle avoidance methods based on artificial potential field are foresee, such as combined with DRL and deep learning.

The domain of research and development concerning mobile robot obstacle avoidance continues to remain an active area of interest. Artificial potential fields (APF) are a common and effective method for obstacle avoidance path planning, where the robot is guided to the target location by a simulated environmental potential field. Traditional artificial potential field methods tend to trap robots in local minima, impeding their ability to reach the goal. This research endeavours to introduce a new approach, the Improved Artificial Potential Field (IAPF) algorithm, which incorporates the A-star method in constructing the artificial potential field. This technique more effectively addresses the issue of path planning for mobile robots, thereby avoiding local minimum solutions. Through simulation experiments in different scenarios, the feasibility of the IAPF algorithm of this paper is verified. The results show that, compared with the traditional APF method, the IAPF algorithm can solve problem of local minimum and plan a sensible path.

This paper presents a novel artificial potential field method for obstacle avoidance and path planning of mobile robots. By analyzing the shortcoming of the artificial potential field methods for robot path planning, we propose an obstacle avoidance method based on gravity chain. Suppose that there is a rubber band which connects with the beginning and the ending in the obstacle potential field space. As the rubber band will be the role of potential field power, we can build a model to simulate the shape of the rubber band. Then this method will generate a steer angle tangent to the rubber band instead of the angle of artificial potential field. By putting effective obstacle avoidance information into potential field through gravity chain, we solve the problems that the artificial potential field method often converges to local minima, as well as it hardly reach the ending and oscillatory movement. The Simulation results show that the method proposed is correct and effective.

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.

The paper discusses the problem of obstacle avoidance of a self-driving car in urban road conditions. The artificial potential field method is used to simulate traffic lanes and cars in a road environment. The characteristics of the urban environment, as well as the features and disadvantages of existing methods based on the structure of planning-tracking, are analyzed. A method of local path planning is developed, based on the idea of an artificial potential field and model predictive control in order to unify the process of path planning and tracking to effectively cope with the dynamic urban environment. The potential field functions are introduced into the path planning task as constraints. Based on model predictive control, a path planning controller is developed, combined with the physical constraints of the vehicle, to avoid obstacles and execute the expected commands from the top level as the target for the task. A joint simulation was performed using MATLAB and CarSim programs to test the feasibility of the proposed path planning method. The results show the effectiveness of the proposed method.

To solve the problems of path planning and path tracking in the field of automatic driving, the path planning and tracking methods applicable to the actual environment under complex constraints are studied. Firstly, the improved double-tree RRT algorithm (RRT-Connect) based on bidirectional extended balance was introduced. Under the advantage of avoiding the modeling of space, the target preference function and the metric function were introduced in combination with the environmental constraints and the constraints of the vehicle itself. At the same time, combined with the regression detection and collision detection mechanism to solve the local minimum problems in motion planning, and greatly improve the effectiveness of path planning. Then generate a smooth continuous executable track based on the cubic spline interpolation function. Finally, the heading angle is obtained according to the vector field method and converted into a tangential angle to continuously track the path. The effectiveness, correctness and practicability of the algorithm are verified by simulation experiments.