International Journal of Vehicle Autonomous Systems (5 papers in press)

Regular Issues

• Visual SLAM and autonomous accompanying flight for UAV transmission line inspection  
  by Hao Wang, Dunbin Xu, Yudong Bai, Yang Zheng, Yuxuan Yao, Zhipeng Xu, Guo Peixin 
  Abstract: This paper addresses UAV transmission-line inspection with a prior-assisted visual SLAM and autonomous following pipeline. The proposed system constructs a robust visual SLAM framework using an Enhanced Multi-scale Line Segment Detector (EM-LSD) and Local Orthogonal Plane Groups (LOPG) constraints, enabling real-time mapping of conductors and towers. Coarse tower coordinates seed the mission, while EM-LSD, LOPG constraints, a semantic-topological map, and bounded visual servoing reduce dependence on dense waypoint trajectories and continuous high-accuracy GNSS during conductor following. A hierarchical controller combining image-based visual servoing (IBVS) and model-based 3D pose control governs decision-making. Real-world experiments demonstrate positioning error below 0.9 m, conductor reconstruction accuracy of 0.3 m, a conductor capture success rate of 95%, and lateral deviation within ±0.5 m over flights exceeding 1 km. The detector benchmark further reports an EM-LSD F-score of 0.86 and a runtime of 17.8 ms per frame on the transmission-line test split.
  Keywords: transmission line inspection; visual SLAM; autonomous accompanying flight; semantic-topological map; line feature extraction.
  DOI: 10.1504/IJVAS.2026.10080806
   
  
• Research on the automobile shock absorber control strategy based on MPC control algorithm and parameter design  
  by Tao Hai 
  Abstract: The semi-active suspension of the vehicle is a typical multiple-input, multiple-output system that is distinguished by its rapid dynamics, strong coupling, and actuator constraints. This paper introduces a model predictive control (MPC) approach to the regulation of the damping force of shock absorbers in these types of systems. While adhering to actuator constraints, the EMPC approach minimises the control objective function. This method's minimal computational demand is a significant advantage, as it is well-suited for implementation on standard automotive microcontrollers. MPC system design entails the development of controllers, simulation-based validation, formulation of the control objective function, and mathematical modelling. The simulation results indicate that the proposed EMPC approach provides superior control performance for vehicle semi-active suspension systems when contrasted with conventional control methods.
  Keywords: MPC; model predictive control; SMC; sliding mode control; pattern search; genetic algorithms; differential evolution.
  DOI: 10.1504/IJVAS.2026.10080592
   
  
• Autonomous navigation and path planning in an unmanned aerial vehicle using model predictive control  
  by Amit Yadav, Lava Bhargava 
  Abstract: In recent decades, unmanned aerial vehicles, especially drones and micro-UAVs, have shown massive improvements in terms of their structure, navigation guidance control, working methodology, and flying features. Autonomous technology is used in different applications such as communication, task scheduling, delivery system, inspection, path planning, etc. The research focuses on the mathematical modelling of the system, in terms of dynamics, optimal control, path planning and trajectory optimisation. Another part is concerned with filtering, system identification, and obstacle avoidance. The mathematical modelling of UAV treating the key models involves enabling control simulation, mapping propeller angular velocities to thrust, torque, path planning, and mapping. It explores the basic principles that guide UAV motion, including six-degree-of-freedom flight dynamics, aerodynamic forces, and actuator mechanisms. The discussion covers different control design methods, from classical PID to Model Predictive Control (MPC) which are important for achieving strong flight autonomy. Autonomous navigation is a fundamental capability for modern aircraft systems, enabling them to operate efficiently and safely in dynamic and uncertain environments.
  Keywords: UAV; path planning; model predictive control; trajectory optimisation; system integration.
  DOI: 10.1504/IJVAS.2026.10080775
   
  
• Analysing the performance and stability of QZS-THS and QZS-TAS on vehicles of PCs, HTs, and SHs  
  by Li Zhang, Yaxi Liu, Vanliem Nguyen, Trungkien Nguyen, Tronghoan Nguyen 
  Abstract: To analyse the isolation performance and stability of QZS-THS (quasi-zero stiffness using two horizontal springs) and QZS-TAS (quasi-zero stiffness using two air springs) in improving the RS (ride smoothness) of PCs (passenger cars), HTs (heavy trucks), and SHs (sunflower harvesters), the models of vehicles and their seat suspension supported by QZS-THS and QZS-TAS are built to analyse the results. The investigation indicates that RS of HTs is better than that of SHs, while RS of PCs is the best under the typical operating conditions of each type of vehicle. When QZS-THS and QZS-TAS are used, the efficiency and stability of QZS-THS and QZS-TAS on PCs and SHs are similar. However, the efficiency and stability of QZS-TAS on HTs are better than those of QZS-THS. Therefore, both QZS-THS and QZS-TAS could be applied to PCs and SHs. Conversely, QZS-TAS should be applied to HTs to optimise its isolation performance.
  Keywords: passenger cars; heavy trucks; sunflower harvesters; quasi-zero stiffness; two horizontal springs; two air springs; ride smoothness.
  DOI: 10.1504/IJVAS.2026.10080591
   
  
• Optimal integration of an autonomous vehicle into a target trajectory considering kinematic constraints  
  by Halima Soundouss, Mohammed Msaaf, Mohamed Fri, Fouad Belmajdoub 
  Abstract: Trajectory planning represents a fundamental challenge in autonomous driving, particularly for ensuring the safe and optimal integration of an autonomous vehicle into a target path. This paper addresses the specific problem of merging a vehicle from an off-path initial state into a predefined trajectory while respecting its non-holonomic and dynamic constraints. We propose a two-stage methodology: first, the minimum feasible turning radius is estimated from a ten-degree-of-freedom (10-DOF) vehicle dynamic model coupled with the Pacejka tyre model, using a neural network trained on simulation-generated data to map driving and road conditions to the corresponding radius; second, a geometric optimisation algorithm, extending our previous geometric trajectory design method, searches among candidate junction points on the target trajectory to identify the optimal insertion point and speed that minimise a weighted cost function combining travelled and remaining distance. Simulation results across multiple initial vehicle states demonstrate that the proposed method reliably identifies feasible and efficient integration trajectories. The main contributions of this work are: (i) a data-driven estimation of the achievable turning radius that accounts for road and dynamic conditions, and (ii) an optimal junction-point search algorithm for trajectory integration under kinematic constraints.
  Keywords: trajectory integration; autonomous vehicle; neural network-based curvature estimation; non-holonomic constraints; vehicle dynamics; optimal path planning; cost function optimisation.
  DOI: 10.1504/IJVAS.2026.10080920