For the RRT* algorithm, there are problems such as greater randomness, longer time consumption, more redundant nodes, and inability to perform local obstacle avoidance when encountering unknown obstacles in path planning process of autonomous vehicles. And artificial potential field method (APF) applied vehicles is prone optimality, unreachable targets, inapplicability global scenarios. A fusion algorithm combining improved proposed. First all, for concept probability sampling optimization...

In view of the fact that global planning algorithm cannot avoid unknown dynamic and static obstacles local easily falls into optimization in large-scale environments, an improved path based on integration A* DWA is proposed applied to driverless ferry vehicles. Aiming at traditional algorithm, vector angle cosine value introduced improve heuristic function enhance search direction; neighborhood expanded optimized efficiency; aiming problem there are many turning points a cubic quasi-uniform...

This study aims to explore an intelligent vehicle trajectory tracking control method based on optimal theory. Considering the limitations of existing strategies in dealing with signal delays and communication lags, a strategy combining anthropomorphic forward-looking reference path longitudinal velocity closure is proposed improve accuracy stability tracking. Firstly, according dynamic error model, linear quadratic regulator (LQR) transverse controller designed principle, feedforward added...

This paper focuses on the obstacle avoidance trajectory planning problem for autonomous vehicles structured roads. The objective is to design a algorithm that can ensure vehicle safety and comfort satisfy rationality of traffic regulations. proposes path speed decoupled method non-split Firstly, layer adopts improved artificial potential field method. obstacle-repulsive field, gravitational fitting traditional are improved. Secondly, aspect performed in Frenet coordinate system. Speed...

Aiming at the problems of A* algorithm’s long running time, large number search nodes, tortuous paths, and planned paths being prone to colliding with corner points obstacles, adaptive weighting reward value theory are proposed improve it. Firstly, diagonal-free five-way based on coordinate changes is used make algorithm purposeful. Meanwhile, in order path security, diagonal filtered out when there obstacles neighborhood. Secondly, a radial basis function act as coefficient heuristic adjust...

Obtaining accurate vehicle driving state and road adhesion coefficient information is of great significance to many aspects the vehicle. This paper takes distributed drive electric vehicles as research object designs a joint estimation method based on theory federated-cubature Kalman filter. The corresponding nonlinear three-degree-of-freedom dynamics model established space equation obtained. Multi-source fusion low-cost sensor signals carried out by using technology, an algorithm estimator...

How to improve the driving performance of vehicle while carrying out path tracking control has become a hot issue in current research. In this paper, an MPC (Model predictive control) algorithm incorporating differential braking is proposed. By establishing dynamics model semi-trailer train, theory adopted for tracking. Then, model, considering additional yaw moment, established design strategy. Under low-speed working conditions, PID (Proportional Integral Derivative) used solve moment with...

Aiming at the problem that traditional rapidly exploring random tree (RRT) algorithm only considers global path of unmanned vehicles in a static environment, which has limitation not being able to avoid unknown dynamic obstacles real time, and window approach (DWA) is prone fall into local optimum during planning, this paper proposes planning method for integrates improved RRT DWA algorithms. The by introducing strategies such as target-biased sampling, adaptive step size, radius node...

This paper addresses the challenge of trajectory planning for autonomous vehicles operating in complex, constrained environments. The proposed method enhances hybrid A-star algorithm through back-end optimization. An adaptive node expansion strategy is introduced to handle varying environmental complexities. By integrating Dijkstra’s shortest path search, improves direction selection and refines estimated cost function. Utilizing characteristics planning, a quadratic programming approach...

Because LiFePO4 batteries are widely used in the energy storage system, their safety has received a great deal of research. If an internal short circuit (ISC) Li-ion battery system leads to thermal runaway, it will pose uncontrollable hazard. Therefore, there is proposed method this study that one-dimensional voltage-correlated convolutional neural network (1DVCNN) detect and localize ISC faults time, which combination feature extraction, correlation analysis, networks. First, voltage signal...

The purpose of this paper is to alleviate the potential safety problems associated with human driver and automatic system competing for right way due different objectives by mitigating human-machine conflict phenomenon in shared driving (HMSD) technology from automation system. Firstly, a basic lane- changing trajectory algorithm based on quintic polynomial Frenet coordinate developed. Then, order make planned close behavior, naturalistic data collected, which some performance features are...

As a new type of transportation, the distributed drive electric vehicle is regarded as main development direction vehicles in future. Due to advantages independently controllable driving torque each wheel, it provides more favorable conditions for active safety control. Acquiring accurate and real-time parameters such speed side slip angle prerequisite Therefore, relying on National Natural Science Foundation China, this paper takes form four-wheel independent steering research object....

An integrated control algorithm based on model predictive (MPC) for vehicle active front steering (AFS) and yaw moment (DYC) is proposed in this paper. Hierarchical structure adopted the algorithm. Design controller use quadric programming method tire force allocation. The verified by simulation. results show that can enhance stability feasible.

Aiming at the problem that it is difficult to ensure trajectory tracking accuracy and driving stability of driverless racing car under extreme conditions high-speed turning with different road adhesion coefficients, a control strategy proposed. Firstly, coefficient estimated using extended Kalman filter algorithm. Draw phase plane diagram vehicle's centroid sideslip angle-centroid angular velocity. Use two-line method determine area obtain expected limit vehicle speed coefficients front...

Autonomous mini-buses are low-cost passenger vehicles that travel along designated routes in industrial parks. In order to achieve this goal, it is necessary implement functionalities such as lane-keeping and obstacle avoidance. To address the challenge of deploying deep learning algorithms detect environmental information on low-performance computing units, which leads difficulties model deployment inability meet real-time requirements, a lightweight algorithm called YOLOP-E based YOLOP...

With the development of science and technology, self-driving technology is gradually being applied to automobile semi-trailer trains. Aiming at problem that it challenging plan traveling trajectory an autonomous train, a planning algorithm based on constrained iterative linear quadratic regulator proposed. The solves cannot deal with inequality constraints by transforming into interior point penalty functions all kinds in problem. When dealing obstacle avoidance problem, body contour modeled...

The purpose of this game theory-based trajectory tracking control study driverless cars is to resolve the conflicting problems precision in and drive stability for lane change situations. general plan made. lateral based on theory evolutionary games, linear quadratic regulator (LQR) a method with predictive feedforward. When it comes dynamic systems that vary over time, are both sides same coin. payoff matrix first constructed determine utility function, followed by dynamical replication...

In driverless formula car racing, cone detection faces two significant challenges: one is recognizing cones at long distances accurately, and the other being prone to leakage under bright light conditions. These challenges directly affect accuracy response speed. order cope with these problems, thesis based on YOLOv8s improve bucket algorithm. Firstly, a P2 layer for detecting tiny objects added top of detect small targets 160 × pixels, which improves conical buckets in distant view. At same...

To solve the uncertainties in permanent magnet synchronous motor (PMSM) driven steer-by-wire (SbW) system, a prescribed-performance-control (PPC) based finite-time fuzzy controller is proposed. Specifically, dynamics of SbW system by PMSM analyzed, and combine PPC technology to transform tracking error coordinates. Then, velocity signal obtained from sigmoid function-based differentiator using position signal. problem mismatched disturbance, logic (FLS) are used as universal approximator...

Vehicle running state parameters and road surface are crucial to the stability of four-wheel independent drive steering electric vehicle control. Therefore, this study explores estimation driving adhesion coefficients using a combination federal Kalman filtering an intelligent bionic antlion optimization algorithm. Firstly, according research purpose paper focus on accuracy establishment three degrees freedom dynamics model, fully considering conditions, adopts Dugoff tire model finally...

Aiming at the problem that it is difficult to ensure trajectory tracking accuracy and driving stability of distributed driven driverless formula racing car under high-speed conditions, a lateral longitudinal coordinated control strategy proposed. Based on adaptive model predictive theory, motion controller designed, prediction time domain changed in real according change vehicle speed. sliding mode variable structure designed accurately track desired Considering coupling between controls,...

The traditional <a:math xmlns:a="http://www.w3.org/1998/Math/MathML" id="M2"> <a:msup> <a:mrow> <a:mi>A</a:mi> </a:mrow> <a:mi>∗</a:mi> </a:msup> </a:math> algorithm, applied to the motion planning of autonomous vehicles, easily causes high computational costs and excessive turning points generated in path. In addition, vehicle cannot track path due unsmooth inflection point. To overcome these potential limitations, an improved <c:math xmlns:c="http://www.w3.org/1998/Math/MathML" id="M3">...

Abstract The 4 DOFS dynamic simulation model of two independent rear wheels drive electric vehicle was built by the Matlab/Simulink software. Asynchronous motors were used in .The motor vector control algorithm and slip ratio differential strategy are put forward this paper. Simulation results showed that torque could be achieved with system good steering stability realized algorithm.