Guidance, Navigation and Control
This paper addresses the formation control problem for ground mobile robot formations and proposes a formation transition method based on dynamic density guidance. To achieve different formation transitions, a centroidal Voronoi tessellations (CVT) formation control algorithm is utilized to avoid collisions during the transition process. By leveraging the properties of the CVT algorithm, a dynamic density is generated by constructing a transition density function between the initial formation density function and the desired density function. The CVT algorithm then guides the robots in the formation to move and complete the transition and reconstruction of the formation. The simulation results demonstrate that, compared to using the desired density function directly, this method not only successfully resolves certain formation transition failures but also reduces the average positional error of the formation during the transition process.
The visual assistance driving system for civil aviation aircraft captures information about the surrounding threat scenario using airborne visual sensors, providing pilots with additional information to aid decision-making. However, the threat objects in the airfield obtained by the optical sensors on the airborne differ significantly in scale, and the computing capacity of the onboard platform is limited. Current methods for object detection do not meet the requirements for visual assistance in driving scenarios. To address this issue, a lightweight multi-scale object detection algorithm based on YOLOv5s is proposed. First, the CA-BIFPN feature fusion network is designed by combining the weighted bidirectional feature pyramid network (BIFPN) with the coordinate attention (CA) attention mechanism, which aims to enhance the feature expression of small objects and to improve the capacity of the model to learn multi-scale objects. Then, the GSConv decoupled detection head is designed to improve object detection accuracy by making classification and regression independent. To enhance the detection speed of the network and enable real-time detection of airfield objects, a cross-level partial lightweight neck module is designed to reduce the additional parameters introduced by the decoupled head. A self-built multi-scale airfield object dataset containing real-world and simulated data from airborne visual sensors from a civil aviation aircraft perspective is established to verify the performance of the proposed algorithm. The experiments conducted on this dataset demonstrate that the detection accuracy of the proposed algorithm surpasses that of faster R-CNN, SSD, and other classic multi-scale object detection algorithms like YOLOv6, YOLOv7, and YOLOX. The achieved mean average precision is 71.40%, which is 4.19% higher than that of YOLOv5s. Furthermore, the detection frame rate achieves 71 frame per second on the simulated airborne computing platform, which satisfies the real-time detection requirements.
This paper addresses the issue of perching maneuver of unmanned aerial vehicles in wind-disturbed environments, by combining the control-oriented sparse identification of nonlinear dynamics with control (SINDYc) method and the imitation deep reinforcement learning (IDRL) control strategy. The study focuses on the design of control strategies for perching maneuvers. First, a training environment for the perching system is established using domain randomization, which incorporates various wind conditions. Then, the SINDYc method is employed to learn sparse models of the perching system offline under different wind conditions, using historical data and a candidate function library, to effectively identify the wind information. Afterwards, the perching control strategy is trained using an IDRL algorithm within the training environment that encompasses multiple wind conditions, resulting in a control strategy for perching in wind-disturbed scenarios. Finally, numerical simulations are conducted to verify the effectiveness of the proposed perching control strategy in wind-disturbed environments.
In this paper, a leader-model-follower matching-based distributed adaptive cooperative control scheme is developed by equivalenting abrupt changing topology to switching topology problem for uncertain heterogeneous multi-agent systems with abrupt changing communication topology to realize leader-follower output consensus. First, local output tracking error is proposed to transform the leader-follower global output consensus problem into the neighboring agents local output consensus problem. Then, the distributed nominal cooperative control design is performed with known system parameters to realize reference model-leader matching and follower-reference model matching, to ensure the leader-follower output consensus with abrupt changing communication topology. Afterwards, the distributed adaptive cooperative controller is studied to realize the asymptotic output tracking of the follower to the leader with unknown parameters under the abrupt change of communication topology. The designed control strategy can ensure the closed loop stabilization of the global agents as well as the leader-follower output consensus with switching topology without relying on global information. Finally, the effectiveness of the designed control scheme is verified by simulation.
To address the issue of multi-body folding wings being subjected to aerodynamic load changes, gravity reversal, and external disturbances during coordinated tilting, an improved active disturbance rejection control (ADRC) strategy based on the combination of hyperbolic tangent function and phase compensation is proposed. Using the Lagrange equation, the aerodynamic load on the unit wing is applied to the center of mass, and the offset of the gravity term with the attitude angle of the body is provided, thereby establishing the dynamic model of the multi-body folding wing. An improved ADRC controller is designed to track the desired angular trajectory of the folding wing, and appropriate controller parameters are selected to suppress joint vibration. The simulation results show that the improved ADRC method can effectively achieve trajectory tracking and joint vibration suppression under variable load conditions, and it demonstrates smaller tracking errors and better dynamic performance and robustness when subjected to external disturbances.
To address the limited endurance of rotorcraft and the restricted flight areas of fixed-wing aircraft, a vertically take-off and landing multi-unit wings tandem configuration is proposed. First, using multi-body kinematics and the lifting-line theory, the aerodynamic characteristics of the centralized deformation are analyzed, and limits of the flight angle of attack during the deformation process are determined. Then, a nonlinear multi-body dynamics model is established using the quasi-Lagrangian equation to comprehensively describe the relative motion characteristics between the unit wings. The flight efficiency of fixed wings and folding wings during the climb process is compared, which verifies the long endurance and good maneuverability of the folding wings. Finally, a modal analysis of the attitude and structural coupling characteristics of the folding wings in steady-level flight is conducted, based on which the cooperative control law of folding and flight is designed to achieve stable control in the process of tilting and folding.
In order that the flight vehicle group could form the expected formation, the “leader-follower” formation control law is investigated. First, the distributed extended state observer (DESO) is designed such that the followers could estimate the virtual leader’s position and velocity. Then, the expected positions of the followers are calculated based on the observer outputs and the nominal formation configuration. A dynamic surface control-based position tracking control law is designed for the followers to track the expected positions. Based on the Lyapunov theory, the stability of the proposed method is proved, while numerical simulations validate the effectiveness. The DESO could estimate both the virtual leader’s position and velocity via only the position observations. The method proposed guarantees that the orientation of the formation is consistent with the direction of the virtual leader’s velocity.
To solve the problem of inadequate perception of autonomous driving in occlusion and over-the-horizon scenarios, a vehicle-road collaborative perception method based on a dual-stream feature extraction network is proposed to enhance the 3D object detection capabilities of traffic participants. Feature extraction networks for roadside and vehicle-side scenes are tailored based on respective characteristics. Since roadside has rich and sufficient sensing data and computational resources, the Transformer structure is used to extract more sophisticated and advanced feature representations. Due to limited computational capability and high real-time demands of autonomous vehicles, partial convolution (PConv) is employed to enhance computing efficiency, and the Mamba-VSS module is introduced for efficient perception in complex environments. Collaborative perception between vehicle-side and roadside is accomplished through the selective sharing and fusion of critical perceptual information guided by confidence maps. By training and testing on DAIR-V2X dataset, the model size of vehicle-side feature extraction network is obtained to be 8.1 MB, and the IoU thresholds of 0.5 and 0.7 correspond to the average accuracy indexes of 67.67% and 53.74%. The experiment verifies the advantages of this method in detection accuracy and model size, and provides a lower-configuration detection scheme for vehicle-road collaboration.
General uninhabited aerial vehicle (UAV) situation assessment methods do not consider the influence of complex external environment on the decision-maker, and usually only get the ranking results of the evaluation. Since the decision-maker needs to make decisions in a short period of time, misjudgments or missing strike windows often occur. To address this problem, a three-way decision model based on the cumulative prospect theory is proposed. First, the method the utilizes intuitionistic fuzzy technique for order preference by similarity to an ideal solution to estimate the conditional probability of each target and obtains the situation assessment result. Next, the method calculates the intuitionistic fuzzy situation information obtained by the UAV based on the cumulative prospect theory, and obtains the corresponding cumulative prospect value when each target performs different actions. Finally, based on the principle of maximizing the cumulative prospect value, a new three-way decision rule is derived to divide the situational assessment results into three regions. The experimental analysis shows that the method not only obtains the target threat ranking, but also classifies the target threat level objectively. At the same time, it considers the psychology of the decision-maker in the assessment process, and obtains the target threat assessment results that meet the traits of the decision-maker, providing a reasonable decision support for the complex and changing air combat.
Aimed at the modeling error of the integrated strapdown inertial navigation system(SINS)/global positioning system (GPS) navigation system and the particle degradation problem of particle filter(PF), an in-flight alignment method of integrated SINS/GPS navigation system based on the combined PF-UKF filter is proposed, in combiation with the unscented Kalman filter(UKF). First, the attitude angle is replaced by the error quaternion. The position and velocity differences between SINS and GPS are selected as the observation variables. In addition, a novel error equation of the integrated navigation system is established. Moreover, the sampled particles are divided into random particles and deterministic particles in the proposed combined PF-UKF filter. The random particles are collected by probability density function, and the determined particles are the state values obtained by collecting sigma point of UKF algorithm. Therefore, the proposed method can effectively reduce the complexity of PF and the degree of particle degradation. The simulation results show that compared with the UKF algorithm, the proposed method can effectively improve the error accuracy of integrated navigation system with a better robustness.
A linear parameter-varying (LPV) integrated control law is designed for a hypersonic vehicle to achieve trajectory control based on an altitude-horizontal trajectory control concept. The LPV output-feedback control theory and pole placement techniques are employed to design parameters of the control law within a Mach number envelope. Such a control law performs integrated control for longitudinal and lateral-directional dynamics of the vehicle, free from the scheme of inner and outer control loops of classical flight controls and ensuring robust and optimal control performance in the sense of L2-induced norm. A mathematical model of the hypersonic vehicle is developed in the Earth-centered-Earth-fixed reference frame. Earth rotation, Earth oblateness, and the second order harmonic perturbations of Earth are considered in the model. Numerical simulations are conducted to examine the performance of the LPV controller. The simulation results indicate that the closed-loop system of the hypersonic vehicle achieves D-stability. The LPV control law achieves a good performance in vehicle trajectory control and has sufficient robustness with respect to perturbations and sensor noise.
When the deep space probe approaches the target planet, due to the rapid increase of the gravity of the target planet, the orbital dynamics model will have a rapid acceleration change. Because the noise covariance is not completely known, the traditional filtering algorithm cannot obtain the optimal estimation of navigation parameters, which is difficult to meet the performance requirements of the approach navigation system. In order to meet the requirements of high stability and accuracy of the system, a sliding window adaptive nonlinear filtering algorithm based on system noise covariance is proposed. By constructing the system noise covariance update function and using the sliding window to smooth the noise covariance, the errors caused by velocity noise and position noise are separated, the filter parameter information used is updated in real time, and the system noise covariance is adjusted adaptively. Taking the Mars probe as an example, the simulation results show that, compared with the traditional unscented Kalman filtering method, the position accuracy and velocity accuracy of the proposed method are improved by 90.97% and 66.17% respectively, which suppresses the rapidly changing integral error on the system model, and solves the divergence problem of the traditional filtering method. In addition, the influence of filtering period and window size on navigation performance is analyzed, which provides a feasible new adaptive filtering method for autonomous navigation of deep space exploration.
Because the navigation errors of inertial navigation system accumulate with time, the unmanned aerial vehicle (UAV) formation that only relies on inertial navigation system for positioning cannot obtain precision navigation information in long time flight. To solve this problem, this paper proposes a cooperative navigation scheme for master-slave UAV formation. First, the UAV is equipped with relative navigation sensors to measure the relative velocity and position information between the members of the master-slave UAV formation. Then, considering the relative pose of formation members, the spatial unified transformation scheme is studied. The absolute navigation information measured by each member of UAV formation by inertial navigation system and the relative navigation information measured by relative sensors is unified into the same navigation coordinate system. Finally, a cooperative navigation scheme based on relative velocity and relative position assistance is given. The 30 min simulation results show that the speed and position errors of each cluster converge to 0.1 m/s and 5 m respectively under this scheme, which is more suitable than the inertial navigation system.
In this paper, a nonlinear control method is proposed based on the framework of gain adaptive sliding mode control to deal with the attitude control problem of an unmanned aerial vehicle (UAV), which shows a strong robustness with respect to dynamical uncertainties and external disturbance. In the proposed method, an adaptive gain schedule scheme is proposed to deal with dynamical uncertainties while suppressing the chattering in the sliding mode control. First, the UAV model is introduced and its mathematical model is given. Then, the error is used as the state variable to design a stably converging sliding mode surface, and the gain adaptive super-twisting sliding mode (ASTSM) algorithm is used to design a UAV attitude controller that can converge in finite time, and the stability of the closed-loop UAV system is demonstrated by the Lyapunov’s second method. Finally, the efficiency of the proposed method is demonstrated through comparative simulations.
Entry footprint is an essential manifestation of vehicle maneuverability, which can provide the basis for trajectory planning and guidance, landing point selection, etc. A fast footprint-generation method based on the pseudospectral method is proposed. The influencing factors of footprint are simulated and analyzed. In this method, the attack and bank angles are simultaneously discretized as control quantities to form the nonlinear programming problem of the pseudospectral method, and the footprint is obtained by solving the maximum transverse range problem for several different longitudinal conditions. Moreover, the affecting factors of the footprint are studied. The simulation results show that the mass, reference area, atmospheric density, etc., do not cause the change of the footprint within a specific range. Beyond a certain range, the short longitudinal trajectory would be significantly affected. The left half of the footprint is affected, while the right half is not changed. The effect of the lift-to-resistance ratio on the footprint is significant, and its size is positively related to the footprint range.
A decoupling controller based on linear/nonlinear switching active disturbance rejection control (SADRC) is designed for a class of continuous linear nominal multi-input multi-output (MIMO) system. The system model is introduced and a linear matrix inequality (LMI) based absolute stability analysis approach is proposed for the designed SADRC decoupling controller. Finally, the effectiveness of the proposed method is verified by numerical simulations.
Guidance, Navigation and Control
