Vehicle Kinematics Research Articles

Advancing Vehicle Trajectory Prediction: A Probabilistic Approach Using Combined Sequential Models

Abstract This paper addresses the critical need to quantify vehicle trajectory uncertainty in autonomous driving under environmental variability. We focus on predicting the posterior distribution of vehicle trajectories over a fixed horizon, given an initial state and a sequence of actions. We propose and compare three approaches: a probabilistic seq2seq model based on stochastic variational Gaussian processes, sequential Monte Carlo simulation with a single-step Gaussian process model, and a hybrid model that leverages the strengths of both methods. Each approach incorporates a baseline vehicle kinematics model to enhance stability and convergence. We evaluate these methods using a dataset generated from the CARLA simulator, assessing both point error metrics and probabilistic prediction metrics. This research introduces novel approaches to quantifying vehicle trajectory uncertainty through various uncertainty quantification techniques, with the goal of improving the safety and reliability of autonomous vehicle control systems.

Responses of Vehicular Occupants During Emergency Braking and Aggressive Lane-Change Maneuvers.

To validate active human body models for investigating occupant safety in autonomous cars, it is crucial to comprehend the responses of vehicle occupants during evasive maneuvers. This study sought to quantify the behavior of midsize male and small female passenger seat occupants in both upright and reclined postures during three types of vehicle maneuvers. Volunteer tests were conducted using a minivan, where vehicle kinematics were measured with a DGPS sensor and occupant kinematics were captured with a stereo-vision motion capture system. Seatbelt loads, belt pull-out, and footrest reaction forces were also documented. The interior of the vehicle was 3D-scanned for modeling purposes. Results indicated that seatback angles significantly affected occupant kinematics, with small female volunteers displaying reduced head and torso movements, except during emergency braking with a upright posture seatback. Lane-change maneuvers revealed that maximum lateral head excursions varied depending on the maneuver's direction. The study concluded that seatback angles were crucial in determining the extent of occupant movement, with notable variations in head and torso excursions observed. The collected data assist in understanding occupant behavior during evasive maneuvers and contribute to the validation of human body models, offering essential insights for enhancing safety systems in autonomous vehicles.

Model predictive control for autonomous vehicle path tracking through optimized kinematics

Tracking performance and stability of path tracking is crucial for unmanned vehicles' navigational tasks. Researches on vehicle path tracking controllers primarily rely on dynamic models. In contrast, there are less designs and researches that focus on path tracking controller based on kinematic model. This scarcity may stem from the perceived inadequacy in the accuracy of vehicle kinematic models. This research introduces a novel method for modifying the traditional vehicle kinematic model by incorporating a front wheel steering angle modification function to enhance tracking performance of path tracking controllers based on kinematic models. In terms of control strategy selection, the study opted for nonlinear model predictive control with terminal cost. A controller which is founded on the modified model was used on a simulated vehicle on CarSim-Simulink platform to assess the tracking performance. The simulation was designed with a double-shift line reference trajectory and the initial speeds of the simulated vehicle are 5 m/s and 10 m/s, respectively. The results of the simulations indicate that employing a controller based on the modified model could reduce the peak tracking error by 76.45 % and 43.06 %, and the root mean square of the tracking error was reduced by 73.29 % and 36.06 %, respectively.

Dynamics of tracked vehicles during nonuniform turning on level terrain and on slopes

This paper presents a detailed kinematic and dynamic model of nonuniform skid steering for tracked vehicles on level terrain and uniform skid steering slopes. The objective is to find the required forces applied on both tracks to achieve the desired turning, aiding in determining the needed engine power for effective turning operation. The vehicle kinematics during turning is presented, followed by a complete dynamic analysis that considers the effect of centrifugal and inertia forces, as well as adhesion with the ground. The dynamic analysis of turning on slopes is developed, where the cases of turning while moving uphill and downhill are considered. Moreover, simulation analysis using MATLAB model is conducted to investigate the main turning characteristics on both level terrain and slopes. The model can accurately predict the required forces on both tracks for effective turning, and can aid in the design of more efficient tracked vehicles. Generally, the findings of this study can inform the design of more efficient tracked vehicles by providing insights into the needed driving forces and the required power to achieve the desired radius of turning at a specific vehicle speed.

Enhancing Planning for Autonomous Driving via an Iterative Optimization Framework Incorporating Safety-Critical Trajectory Generation

Ensuring the safety of autonomous vehicles (AVs) in complex and high-risk traffic scenarios remains a critical unresolved challenge. Current AV planning methods exhibit limitations in generating robust driving trajectories that effectively avoid collisions, highlighting the urgent need for improved planning strategies to address these issues. This paper introduces a novel iterative optimization framework that incorporates safety-critical trajectory generation to enhance AV planning. The use of the HighD dataset, which is collected using the wide-area remote sensing capabilities of unmanned aerial vehicles (UAVs), is fundamental to the framework. Remote sensing enables large-scale real-time observation of traffic conditions, providing precise data on vehicle dynamics, road structures, and surrounding environments. To generate safety-critical trajectories, the decoder within the conditional variational auto-encoder (CVAE) is innovatively designed through a data-mechanism integration method, ensuring that these trajectories strictly adhere to vehicle kinematic constraints. Furthermore, two parallel CVAEs (Dual-CVAE) are trained collaboratively by a shared objective function to implicitly model the multi-vehicle interactions. Inspired by the concept of “learning to collide”, adversarial optimization is integrated into the Dual-CVAE (Adv. Dual-CVAE), facilitating efficient generation from normal to safety-critical trajectories. Building upon this, these generated trajectories are then incorporated into an iterative optimization framework, significantly enhancing the AV’s planning ability to avoid collisions. This framework decomposes the optimization process into stages, initially addressing normal trajectories and progressively tackling more safety-critical and collision trajectories. Finally, comparative case studies of enhancing AV planning are conducted and the simulation results demonstrate that the proposed method can efficiently enhance AV planning by generating safety-critical trajectories.

PB-Trajectron: Physics bounded neural network for generalized trajectory prediction

Vehicle trajectory prediction is a critical technology in autonomous driving systems, as the quality of prediction directly affects downstream path planning and vehicle control. Although recent studies have shown that models combining kinematic rules with data-driven methods exhibit better interpretability in trajectory prediction, these models often adopt fixed constraint strengths, limiting their generalization ability in diverse traffic scenarios. This fixed constraint strength design restricts the model’s adaptability to different traffic environments.To address this issue, we propose PB-Trajectron, a dynamic integration mechanism that enhances the model’s prediction performance and generalization capability. PB-Trajectron is a dynamic integration mechanism that combines vehicle kinematic rules with deep learning prediction models for generalized vehicle trajectory prediction. The model integrates physics-based motion models and Deep Neural Networks (DNNs) to provide reasonable physical explanations for predicting vehicle trajectories at different speeds. First, we establish two vehicle kinematic constraints applicable to different prediction scenarios, enabling the agent to switch between these constraints based on the causal relationships between vehicle motion states to avoid generating non-physical trajectory results. Second, we propose a kinematic constraint decision framework based on velocity thresholds, which demonstrates adaptive adjustment, allowing the agent to switch tasks according to real-time state conditions and adaptively adjust the contribution of different physical constraints based on actual vehicle speeds. Finally, we investigate the advantages of PB-Trajectron in Out-of-Domain(OOD) generalization.Cross-scenario experiments on the nuScenes dataset show that, compared to previous methods, PB-Trajectron reduces the final displacement error by 7.69% when the prediction step length is 4. Furthermore, out-of-domain generalization tests on the INTERACTION dataset demonstrate that PB-Trajectron achieves a 12.23% reduction in average prediction error on datasets from different countries and scenarios compared to previous methods. The proposed mechanism can better adapt to complex and diverse traffic scenarios, laying the foundation for explainable and robust autonomous driving systems.

A heavy-duty tracked vehicle model with a reduced feasible domain for motion tracking control considering dynamic characters of hybrid powertrain

Motion tracking control is an essential part of heavy-duty unmanned tracked vehicles to realize autonomous mobility, which adjusts the vehicle motion in real-time by controlling the driving torques. Achieving accurate motion tracking strongly depends on the use of a precise vehicle dynamic model. To support the motion tracking ability, a vehicle dynamic model with a reduced feasible domain is proposed for heavy-duty unmanned tracked vehicles. Firstly, a vehicle dynamic model is established based on the vehicle kinematics and dynamics. Building upon the established vehicle dynamic model, a graph neural network (GNN) is employed to reduce the feasible domain of the vehicle dynamic model, specifically focusing on the limitation of the powertrain. The powertrain components and their interconnections are abstracted into a graph, which is analyzed to study the interactions between the components using the GNN. Finally, the proposed vehicle dynamic model with a reduced feasible domain is verified through actual heavy-duty tracked vehicle experiments in various cases. Compared with the back propagation neural network, the evaluation error of the driving torque decreased by 25.86%. Results show that the proposed vehicle dynamic model with a reduced feasible domain is closer to the performance of the actual heavy-duty tracked vehicle and their powertrain. And the vehicle dynamic model accurately reduces the solution time of the motion tracking control, resulting in improved accuracy and efficiency of the motion tracking control.

User comfort and naturalness of automated driving: The effect of vehicle kinematic and proxemic factors on subjective response

User comfort in higher-level Automated Vehicles (AVs, SAE Level 4+) is crucial for public acceptance. AV driving styles, characterised by vehicle kinematic and proxemic factors, affect user comfort, with “human-like” driving styles expected to provide natural feelings. We investigated a) how the kinematic and proxemic factors of an AV's driving style affect users' evaluation of comfort and naturalness, and b) how the similarities between automated and users' manual driving styles affect user evaluation.Using a motion-based driving simulator, participants experienced three Level 4 automated driving styles: two human-like (defensive, aggressive) and one machine-like. They also manually drove the same route. Participants rated their comfort and naturalness of each automated controller, across twenty-four varied UK road sections. We calculated maximum absolute values of the kinematic and proxemic factors affecting the AV's driving styles in longitudinal, lateral, and vertical directions, for each road section, to characterise the automated driving styles. The Euclidean distance between AV and manual driving styles, in terms of kinematic and proxemic factors, was calculated to characterise the human-like driving style of the AV.We used mixed-effects models to examine a) the effect of AV's kinematic and proxemic factors on the evaluation of comfort and naturalness, and b) how similarities between manual and automated driving styles affected the evaluation. Results showed significant effects of lateral and rotational kinematic factors on comfort and naturalness, with longitudinal kinematic factors having a less prominent effect. Similarities in vehicle metrics, such as speed, longitudinal jerk, lateral offset, and yaw, between manual and automated driving styles, enhanced user comfort and naturalness.This research facilitates an understanding of how control features of AVs affect user experience, contributing to the design of user-centred controllers and better acceptance of higher-level AVs.

Naturalistic Observations of Human Driving Perceptions and Vehicle Kinematics at Stop Sign-Controlled Intersections

Naturalistic Observations of Human Driving Perceptions and Vehicle Kinematics at Stop Sign-Controlled Intersections

Impact of Texting-Induced Distraction on Driving Behavior Based on Field Operation Tests

Driver distraction has proven to be among the top contributory factors in road crashes worldwide. The involvement of drivers in secondary activities, such as phone usage, conversation with co-passengers, reaching for objects, eating, and so forth, have been shown to be significant causes of in-vehicle driver distraction. Text messaging using mobile phones while driving can be detrimental to the primary driving task as it invokes manual, visual, and cognitive distractions. Numerous studies in the past have analyzed the effect of texting on driver distraction using driving simulators. This study followed a field operation test–based approach to simulate real road conditions and evaluate the impact of texting on driving behavior. The study was conducted at the connected autonomous vehicle testbed at the Indian Institute of Technology Hyderabad, using an instrumented test vehicle. Data on gaze parameters and vehicle kinematics were recorded using eye-tracker glasses and a high-end global positioning system (GPS) data logger. Thirty-one participants drove on the pre-defined path at the testbed under baseline and texting conditions. The findings demonstrated that the texting task significantly affected fixation count, mean speed, and lateral acceleration. The texting task led to a reduction in mean fixation counts, mean speed, and mean lateral acceleration by 44.37%, 34%, and 46.72%, respectively. The age and driving experience of the participants had a significant effect on their behavior. The results indicate the critical potential of texting-based driver distraction and suggest the enforcement of stringent rules for texting while driving to create a safer road environment.

Driver Behavior in Response to Forward Collision Warnings Considering Driving Context

Forward collision warnings (FCW) are designed to warn drivers of potential collisions, but their effectiveness may vary based on driving conditions and driver responses. This study investigated driver behavior in response to real-world FCW alerts across various driving contexts with the aims of evaluating the prevalence and characteristics of FCW events, examining how often these events occur, their severity, and the contexts in which they happen. The current analysis considered driver responses to FCW events, including changes in vehicle kinematics, visual attention, and hand-on-wheel (HOW) behavior. Findings suggest that lower-severity FCWs are more common in local road contexts and associated with adjacent vehicles, while higher-severity FCWs were more frequent on highways and in the presence of lead vehicles. The study highlights the potential value of integrating additional contextual information into FCW systems to inform drivers better and reduce unnecessary alerts.

Application of Atmospheric Augmentation for PPP-RTK with Instantaneous Ambiguity Resolution in Kinematic Vehicle Positioning

The long convergence time and non-robust positioning accuracy are the main factors limiting the application of precision single-point positioning (PPP) in kinematic vehicle navigation. Therefore, a dual/triple-frequency multi-constellation PPP-RTK method with atmospheric augmentation is proposed to achieve cm-level reliable kinematic positioning. The performance was assessed using a set of static station and kinematic vehicle positioning experiments conducted in Wuhan. In the static experiments, instantaneous convergence within 1 s and centimeter-level positioning accuracy were achieved for PPP-RTK using dual-frequency observation. For the kinematic experiments, instantaneous convergence was also achieved for dual-frequency PPP-RTK in open areas, with RMS of 2.6 cm, 2.6 cm, and 7.5 cm in the north, east, and up directions, respectively, with accuracy similar to short-baseline real-time kinematic positioning (RTK). Horizontal positioning errors of less than 0.1 m and 3D positional errors of less than 0.2 m were 99.54% and 98.46%, respectively. Additionally, after the outage of GNSS and during satellite reduction in obstructed environments, faster reconvergence and greater accuracy stability were realized compared with PPP without atmospheric enhancement. Triple-frequency PPP-RTK was able to further enhance the robustness and accuracy of positioning, with RMS of 2.2 cm, 2.0 cm, and 7.3 cm, respectively. In summary, a performance similar to RTK was achieved based on dual-frequency PPP-RTK, demonstrating that PPP-RTK has the potential for lane-level navigation.

Investigating Implicit and Explicit Communication of Highly Automated Vehicles in Japan: How do Light-band eHMIs affect Pedestrians’ Willingness to Cross, Trust and Perceived Safety?

In the near future, pedestrians will face highly automated vehicles on the roads. Highly automated vehicles (HAVs) should have safety-enhancing communication tools to guarantee traffic safety, e.g., vehicle kinematics and external human–machine interfaces (eHMIs). Pedestrians, as highly vulnerable road users, depend on communication with HAVs. Miscommunication between pedestrians and HAVs could quickly result in accidents, and this, in turn, could cause severe impairments for pedestrians. Light-band eHMIs have the potential to enhance traffic safety. However, eHMIs have been less explored in Japan so far. As a first-time approach, this experimental online study shed light on the effect of a light-band eHMI on Japanese pedestrians (N=99). In short video sequences, the participants interacted with two differently sized HAVs equipped with light-band eHMI. We investigated the effect of vehicle size (small vs. large), eHMI status (no eHMI vs. static eHMI vs. dynamic eHMI), and vehicle kinematics (yielding vs. non-yielding) on pedestrians’ willingness to cross, trust, and perceived safety. To investigate possible side effects of eHMIs, we also included experimental conditions in which the eHMI mismatched the vehicle’s kinematics. Results revealed that Japanese were more willing to cross the street and indicated higher trust- and safety ratings when they received information about the vehicle’s intention and automation status (dynamic eHMI) compared to when they received no information (no eHMI) or only about the vehicle automation status (static eHMI). Surprisingly, Japanese participants tended to rely on the eHMI when there was mismatching information between eHMI and vehicle kinematics. Overall, we concluded that light-band eHMIs could contribute to a safe future interaction between pedestrians and HAVs in Japan under the requirement that the eHMI is in accordance with vehicle kinematics.

Localization robustness improvement for an autonomous race car using multiple extended Kalman filters

In this paper, we introduce a vehicle localization method designed for the SZEnergy race car, which competes in the Shell Eco-marathon. The proposed method comprises four different extended Kalman filter-based localization algorithms and a selection algorithm that determines the most suitable one based on vehicle speed, GNSS availability, and signal quality. The low-speed Kalman filters are based on a kinematic vehicle model while the high-speed variants are based on a dynamic vehicle model. Several measurements were performed during test maneuvers to evaluate the performance of the filters. The proposed method succesfully handles sensor miscalibration and GNSS outages.

Experimental study on the water entry of a vehicle with a single canard wing at an initial angle of attack

The use of flow control techniques is important to enhance vehicle motion stability during water entry. This paper introduces a method that applies a canard wing to the water entry problem, and the vertical water-entry process of a vehicle with a single canard wing at an initial angle of attack was experimentally studied. The cavity evolution and vehicle kinematics during water entry at various impact velocities and initial angles of attack were investigated. The results indicate that the fore-end cavity and the attached cavity on the wing are features of a vehicle with a canard wing during water entry. Unlike a wingless vehicle, the canard-wing configuration provides a restoring moment that corrects the deflection of the trajectory and attitude. As the impact velocity increased, the vehicle trajectory gradually corrected to a vertical straight-line trajectory before deflecting in the direction opposite to the side without a wing. As the initial angle of attack increased, the vehicle trajectory gradually shifted towards the side with the wing; however, the attitude remained stable without continuous deflection. The canard wing prevents a continuous change in attitude deflection caused by the initial angle of attack.

Multi-mode vehicle pose estimation under different GNSS conditions

The integrated navigation system combining the global navigation satellite system (GNSS) and inertial navigation system (INS) is a crucial method for pose estimation in the field of autonomous driving technologies. Nevertheless, the accuracy of pose estimation is severely compromised when GNSS signals are obstructed or disrupted. To address this issue, this study introduces a multi-mode pose estimation framework designed to ensure accurate pose estimation even under unstable GNSS conditions. By integrating vehicle kinematics model that considers steering characteristics (VKMSC) and the convolutional neural network-long short-term memory (CNN-LSTM) neural network (NN) model into various estimation modes, the framework enhances the robustness of the integrated navigation system against signal interference. The system dynamically selects the optimal estimation strategy based on the degree of GNSS signal disruption. The proposed method has been validated through real-vehicle experiments, which demonstrate its efficacy in providing precise pose estimation across a spectrum of interference scenarios. Under the multipath and non-line-of-sight (MP/NLOS) mode, compared to the integrated navigation system and the fusion of traditional vehicle kinematic models, the proposed method improved positional estimation accuracy by 61.8 % and 19.7 %, respectively. In GNSS outage mode, the proposed method increased the estimation accuracy by 36.5 % and 12.0 %, respectively, compared to the INS navigation system assisted by the VKMSC and CNN-LSTM network model. The proposed method effectively reduces pose estimation errors in the integrated navigation system during interference and suppresses data fluctuations, thereby enhancing the system's precision and robustness.

Path Planning Algorithm of Orchard Fertilization Robot Based on Multi-Constrained Bessel Curve

Path planning is the core problem of orchard fertilization robots during their operation. The traditional full-coverage job path planning algorithm has problems, such as being not smooth enough and having a large curvature fluctuation, that lead to unsteady running and low working efficiency of robot trajectory tracking. To solve the above problems, an improved A* path planning algorithm based on a multi-constraint Bessel curve is proposed. First, by improving the traditional A* algorithm, the orchard operation path can be fully covered by adding guide points. Second, according to the differential vehicle kinematics model of the orchard fertilization robot, the robot kinematics constraint is combined with a Bessel curve to smooth the turning path of the A* algorithm, and the global path meeting the driving requirements of the orchard fertilization robot is generated by comprehensively considering multiple constraints such as the minimum turning radius and continuous curvature. Finally, the pure tracking algorithm is used to carry out tracking experiments to verify the robot’s driving accuracy. The simulation and experimental results show that the maximum curvature of the planned trajectory is 0.67, which meets the autonomous operation requirements of the orchard fertilization robot. When tracking the linear path in the fertilization area, the average transverse deviation is 0.0157 m, and the maximum transverse deviation is 0.0457 m. When tracking the U-turn path, the average absolute transverse deviation is 0.1081 m, and the maximum transverse deviation is 0.1768 m, which meets the autonomous operation requirements of orchard fertilization robots.

Research on Autonomous Vehicle Path Planning Algorithm Based on Improved RRT* Algorithm and Artificial Potential Field Method.

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 the path planning process of autonomous vehicles. And the artificial potential field method (APF) applied to autonomous vehicles is prone to problems such as local optimality, unreachable targets, and inapplicability to global scenarios. A fusion algorithm combining the improved RRT* algorithm and the improved artificial potential field method is proposed. First of all, for the RRT* algorithm, the concept of the artificial potential field and probability sampling optimization strategy are introduced, and the adaptive step size is designed according to the road curvature. The path post-processing of the planned global path is carried out to reduce the redundant nodes of the generated path, enhance the purpose of sampling, solve the problem where oscillation may occur when expanding near the target point, reduce the randomness of RRT* node sampling, and improve the efficiency of path generation. Secondly, for the artificial potential field method, by designing obstacle avoidance constraints, adding a road boundary repulsion potential field, and optimizing the repulsion function and safety ellipse, the problem of unreachable targets can be solved, unnecessary steering in the path can be reduced, and the safety of the planned path can be improved. In the face of U-shaped obstacles, virtual gravity points are generated to solve the local minimum problem and improve the passing performance of the obstacles. Finally, the fusion algorithm, which combines the improved RRT* algorithm and the improved artificial potential field method, is designed. The former first plans the global path, extracts the path node as the temporary target point of the latter, guides the vehicle to drive, and avoids local obstacles through the improved artificial potential field method when encountered with unknown obstacles, and then smooths the path planned by the fusion algorithm, making the path satisfy the vehicle kinematic constraints. The simulation results in the different road scenes show that the method proposed in this paper can quickly plan a smooth path that is more stable, more accurate, and suitable for vehicle driving.

Optimization‐based motion planning for autonomous agricultural vehicles turning in constrained headlands

AbstractHeadland maneuvering is a crucial part of the field operations performed by autonomous agricultural vehicles (AAVs). While motion planning for headland turning in open fields has been extensively studied and integrated into commercial autoguidance systems, the existing methods primarily address scenarios with ample headland space and thus may not work in more constrained headland geometries. Commercial orchards often contain narrow and irregularly shaped headlands, which may include static obstacles, rendering the task of planning a smooth and collision‐free turning trajectory difficult. To address this challenge, we propose an optimization‐based motion planning algorithm for headland turning under geometrical constraints imposed by headland geometry and obstacles. Our method models the headland and the AAV using convex polytopes as geometric primitives, and calculates optimal and collision‐free turning trajectories in two stages. In the first stage, a coarse path is generated using either a classical pattern‐based turning method or a directional graph‐guided hybrid A* algorithm, depending on the complexity of the headland geometry. The second stage refines this coarse path by feeding it into a numerical optimizer, which considers the vehicle's kinematic, control, and collision‐avoidance constraints to produce a feasible and smooth trajectory. We demonstrate the effectiveness of our algorithm by comparing it to the classical pattern‐based method in various types of headlands. The results show that our optimization‐based planner outperforms the classical planner in generating collision‐free turning trajectories inside constrained headland spaces. Additionally, the trajectories generated by our planner respect the kinematic and control limits of the vehicle and, hence, are easier for a path‐tracking controller to follow. In conclusion, our proposed approach successfully addresses complex motion planning problems in constrained headlands, making it a valuable contribution to the autonomous operation of AAVs, particularly in real‐world orchard environments.

Global path planning for autonomous construction vehicles in building construction: A comparative study with a focus on vehicle kinematic characteristics

Technological advancements in robotics and automation have led to a growing emphasis on integration into construction industry from a global research context. In particular, autonomous construction vehicles (ACVs) stand out as a pivotal innovation. One key aspect of successful deployment of ACVs on construction sites is the path planning technology, especially the global path planning (GPP) methods. Thus, this study conducted a comparative study of the six GPP algorithms from various underlying principles, including graph search methods (e.g. Dijkstra and A*), sampling-based methods (e.g. rapidly exploring random tree (RRT) and rapidly exploring random tree star (RRT*)), evolutionary computation (e.g. genetic algorithms (GAs)), and artificial intelligence (e.g. deep Q learning (DQN)), to reveal how GPP algorithms perform on the travel of ACVs. These algorithms were evaluated with consideration of the kinematic characteristics of typical construction vehicles, including truck-trailers and excavators, within a typical unstructured real-world building construction site and based on two tasks and eight criteria encompassing efficiency and safety. The findings were analysed from three perspectives, i.e. the planned global paths and travel paths of ACVs, the performance of algorithms in each metric, and the overall performance of each algorithm, to identify the strengths of each GPP algorithm. GAs excel in generating smoother travel paths than others, whereas the global paths from RRT and DQN exhibit unpredictable changes, leading to a meandering travel paths for ACVs. The quality of global paths induces a greater impact on differential-wheeled excavators compared to front-wheel steering truck-trailers. Practical guidance can therefore be provided to select and improve the GPP algorithms and thereby to enhance the safety and efficiency of ACVs on construction sites.