Autonomous Guidance Research Articles

Electro-Optical Relative Navigation for Inspection and Rendezvous in the SpEye Mission

The execution of autonomous in-orbit servicing operations is expected to play a crucial role to preserve and enhance satellites’ performance, as well as to ensure a sustainable use of outer space. In this respect, several research efforts are dedicated to addressing the technological challenges that characterize the development of an autonomous guidance, navigation, and control system. In this framework, the Space Eye (SpEye) mission, funded by the Italian Space Agency, aims to demonstrate safe and inexpensive on-orbit inspection of a target satellite using a CubeSat-based free-flying nanosatellite. This paper, which stems from the phase A mission study, describes the high-level architectural design of its electro-optical-based relative navigation subsystem, employing multisensor algorithmic approaches based on the combination of visual, laser, and Differential Global Navigation Satellite System measurements. A dedicated environment is developed for the simulation of measurements from the relative navigation suite (including a synthetic image generator), allowing a preliminary performance assessment of the proposed algorithmic approaches.

Path-tracking Robust Model Predictive Control of an Autonomous Steering System Using LMI Optimization With Independent Constraints Enforcement

Path-tracking Robust Model Predictive Control of an Autonomous Steering System Using LMI Optimization With Independent Constraints Enforcement

Intelligent System for Artillery Mount Autonomous Guidance And Firing Adjustment

The article presents mathematical models, numerical algorithms and a software package for an intelligent system for artillery mount autonomous guidance and adjustment of firing. The software package includes the following components: a module for solving the direct problem of external ballistics, a neural network training module, and a firing guidance and adjustment module. The first module is designed to form a database of computational experiments by repeatedly solving the direct problem of external ballistics under varying firing conditions. In the second module, a neural network is trained to solve the inverse problem of external ballistics using the database generated from computational experiments. The third module implements algorithms for calculating the artillery mount launching angles on a target using a trained neural network and calculating corrections to launching angles based on data on the projectile impact point deviation from the target. Having received data on the artillery mount readiness to fire, the coordinates of the target position and projectile’s impact point received from a digital observer, the software package calculates the necessary launching angles or corrections to them. Launching angles are transferred via text files to the control system of the artillery mount drives, then a shot is fired and the conditions for hitting the target are assessed. The algorithm is then repeated for the next shot or new target. Testing of the developed algorithms for autonomous guidance and shooting correction was carried out using modeling of external ballistics processes in the software package, as well as carrying out experimental testing on a working simulation device of an artillery mount. The developed software package allows visualization of the obtained results and saving the history of the performed operations.

Trajectory optimization of spacecraft autonomous far-distance rapid rendezvous based on deep reinforcement learning

This paper investigates the application of Deep Reinforcement Learning (DRL) in the trajectory optimization of spacecraft far-distance rapid rendezvous, and uses the most advanced DRL method Proximal Policy Optimization (PPO) to solve the continuous high-thrust minimum-fuel trajectory optimization problem. The space J2 perturbation was considered, its impact on the spacecraft’s on-orbit operation and trajectory design was analyzed, and the effectiveness and accuracy of the proposed method were verified in two far-distance rapid rendezvous missions. In order to ensure the safety of the subsequent close-range operation phase, a safe area reward framework is proposed, and sparse and dense safe area reward functions are designed. The dense safe area reward function significantly improves the training efficiency of the algorithm on the basis of ensuring terminal performance. In addition, the modeling and analysis of possible uncertainties in the spacecraft’s orbit operation, including observation uncertainty, state uncertainty and control uncertainty, is carried out to verify the performance of the proposed method through simulation. For uncertainties, the closed-loop performance of the policy is also evaluated by performing Monte Carlo simulations. The results show that the PPO algorithm can effectively deal with the rendezvous problem in uncertainty environments. These preliminary results demonstrate the great potential of the DRL method in achieving autonomous guidance of spacecraft.

On-board radar complex for high-precision direction finding of radio sources for autonomous guidance of winged UAVs

The subject of this article is the design of a cheap, simple, and highly accurate on-board radar system for determining the angular position of radio-signal emission stations, which can be used to guide UAVs to a specified area of space. The aim of this study is to develop high-precision functioning algorithms, practical recommendations for implementation, and experimental measurements of the main parameters of the onboard radio direction finders of radio radiation sources placed on winged UAVs. Objectives of the research: 1) analysis of the statistical theory of optimization of signal processing algorithms in radar systems; 2) synthesis of direction-finding algorithms for radio measurement sources capable of performing measurements with high accuracy and in a wide range of unambiguous measurement angles; 3) synthesis of the radio direction finder structural scheme and its simulation modelling; 4) design of radio direction finder receiving antennas capable of operating separately and in a complex; 5) justification of the choice of components for the input paths of receivers, parameters of ADC and microcomputers that make up the main components of the radar complex; 6) manufacturing of a working model of the radar complex; 7) experimental measurements in the laboratory. The methods for solving the tasks are based on the statistical theory of optimization of radio engineering systems for remote sensing and radar, existing software tools for simulation modelling, and the theory of radio measurements. The main idea behind creating a cheap, simple, and highly accurate on-board radar system for radio emitter direction finding is to combine the results of angular position measurements from several amplitude and phase meters. This approach allows us to create an easy-to-implement radio system with the advantages of different measuring instruments. The following results were obtained: 1) theoretical studies and simulation modelling confirmed that by combining measurements from a two-antenna amplitude direction finder with narrow diagrams, a two-antenna direction finder with wide diagrams, and a two-antenna phase finder, it is possible to achieve both high accuracy and a wide range of unambiguous measurements; 2) a radar system based on a cheap and commonly available element base was developed; 3) antennas were developed that can be easily installed under the wings of UAVs; 4) experimental studies confirmed the performance of complex signal processing in an on-board six-antenna radio direction finder. The material of this work forms the basis for further experimental development of radio-direction finders for various purposes, opens up opportunities for overcoming the contradiction between accuracy and range of unambiguous measurements, and highlights an additional direction for increasing the autonomy of winged UAVs.

Low-energy Earth–Moon transfer autonomous guidance considering high-fidelity orbital dynamics

Low-energy Earth–Moon transfer autonomous guidance considering high-fidelity orbital dynamics

SIMULATION OF AN AUTONOMOUS HYDROSTATIC DRIVE

The features of the application of autonomous electrohydraulic drives for power engineering are considered. The classification of linear actuators is given, the analysis of design and layout solutions of developers of autonomous steering actuators for general engineering is performed. The analysis of disadvantages is carried out and the directions of improvement of autonomous drives are presented. A scheme of an autonomous electrohydraulic drive with an automatic power control system based on the use of information from pressure, flow and displacement sensors is proposed. A mathematical model of an autonomous electrohydraulic drive with a power regulator has been developed, which makes it possible to estimate a random change in the load on the executive hydraulic motor and optimize the pump supply in accordance with the required task. The obtained research results confirm that an autonomous hydraulic drive with a power regulator is a promising technical solution that contributes to the development of high-precision power drives and controllability for general mechanical engineering.

Development of a Compact Transport Vehicle for Autonomous Steering and Tracking Driving in an Open Field Environment

Development of a Compact Transport Vehicle for Autonomous Steering and Tracking Driving in an Open Field Environment

Performance evaluation of rebound damping of target marker

For small body missions, autonomous guidance is crucial for landing. The Japan Aerospace Exploration Agency’s Hayabusa and Hayabusa2 spacecraft thus placed a target marker on an asteroid’s surface for approach and landing maneuvers. Because of the microgravity environment of an asteroid, even a low-velocity impact can result in a high rebound. In order to reduce rebound motion, a target marker is composed of a shell and a large number of inner particles. The inner particles collide with each other in the shell, reducing the coefficient of restitution of the whole system. Since the settling position of the target marker is directly related to landing accuracy, understanding the target marker’s coefficient of restitution is crucial. This work explains the damping mechanism and develops a theory that can be used to evaluate damping performance based on the number and radius of inner particles. The proposed model is validated by a comparison with the results of a numerical simulation based on the discrete element method. The results of this study will contribute to future small body missions, whose success can depend on mitigating rebound on an asteroid surface. Since particle dampers can be applied to a wide variety of payloads, the results of this study are useful for designing separable payloads for future small body missions.

Coordinated torque control for enhanced steering and stability of independently driven mobile robots

PurposeMobile robots with independent wheel control face challenges in steering precision, motion stability and robustness across various wheel and steering system types. This paper aims to propose a coordinated torque distribution control approach that compensates for tracking deviations using the longitudinal moment generated by active steering.Design/methodology/approachBuilding upon a two-degree-of-freedom robot model, an adaptive robust controller is used to compute the total longitudinal moment, while the robot actuator is regulated based on the difference between autonomous steering and the longitudinal moment. An adaptive robust control scheme is developed to achieve accurate and stable generation of the desired total moment value. Furthermore, quadratic programming is used for torque allocation, optimizing maneuverability and tracking precision by considering the robot’s dynamic model, tire load rate and maximum motor torque output.FindingsComparative evaluations with autonomous steering Ackermann speed control and the average torque method validate the superior performance of the proposed control strategy, demonstrating improved tracking accuracy and robot stability under diverse driving conditions.Research limitations/implicationsWhen designing adaptive algorithms, using models with higher degrees of freedom can enhance accuracy. Furthermore, incorporating additional objective functions in moment distribution can be explored to enhance adaptability, particularly in extreme environments.Originality/valueBy combining this method with the path-tracking algorithm, the robot’s structural path-tracking capabilities and ability to navigate a variety of difficult terrains can be optimized and improved.

Adaptive Advanced Emergency Braking on Combined Road Friction Coefficients

In this study, an Advanced Emergency Braking System (AEBS) is adopted to the combined road friction coefficients by proposing an emergency steering maneuver. AEBS is one of the significant driver assistant systems to avoid rear-end collisions. AEBS supports the drivers with acoustic and optical warnings before applying the full braking maneuver. However, in special cases, such as the sudden decrease of road friction coefficient during braking, a single hard braking maneuver without a proper steering assist may not be sufficient to prevent a rear-end collision. Therefore, AEBS may work with the autonomous emergency steering systems to prevent inevitable rear-end collisions. The originality of this study arises from the necessity of the adaptation of the AEBS to work in cooperation with the autonomous emergency steering systems. A predictive controller was used to support the emergency braking maneuver with an autonomous steering maneuver via observing the yaw rate of the vehicle. The predictive controller was developed in MATLAB software. The implementation of the controller was performed in a non-linear four-wheel vehicle model in the IPG/CarMaker simulation environment. The communication between the simulation environment and MATLAB software was established in the Simulink interface of MATLAB. The results proved that the predictive controller maintained the vehicle lateral stability without causing any type of collisions.

Fault detection and fault-tolerant control of dual-motor autonomous steering system

In order to improve the safety of autonomous ground vehicle (AGV) and the accuracy of trajectory tracking, a fault detection (FD) scheme of steering system and a fault-tolerant control (FTC) method are proposed. First, this paper analyzes the causes of steering motor failure and designs a passive fault-tolerant controller to ensure that the torque motor under coordinated control can still follow the angle when a slight fault occurs in the angle motor. Second, the corresponding FD schemes are designed for different motor faults, and the unknown input observer (UIO) of the steering resistance moment is designed considering the uncertainty of the steering system caused by the fault motor. The active fault-tolerant controller composed of backstepping control and global sliding mode control (SMC) is further combined to control the fault-free motor. Finally, simulation and HiL test results show the effectiveness of the proposed FD scheme and FTC strategy.

Design and Characterization of a Powered Wheelchair Autonomous Guidance System.

The current technological revolution driven by advances in machine learning has motivated a wide range of applications aiming to improve our quality of life. Representative of such applications are autonomous and semiautonomous Powered Wheelchairs (PWs), where the focus is on providing a degree of autonomy to the wheelchair user as a matter of guidance and interaction with the environment. Based on these perspectives, the focus of the current research has been on the design of lightweight systems that provide the necessary accuracy in the navigation system while enabling an embedded implementation. This motivated us to develop a real-time measurement methodology that relies on a monocular RGB camera to detect the caregiver's feet based on a deep learning method, followed by the distance measurement of the caregiver from the PW. An important contribution of this article is the metrological characterization of the proposed methodology in comparison with measurements made with dedicated depth cameras. Our results show that despite shifting from 3D imaging to 2D imaging, we can still obtain comparable metrological performances in distance estimation as compared with Light Detection and Ranging (LiDAR) or even improved compared with stereo cameras. In particular, we obtained comparable instrument classes with LiDAR and stereo cameras, with measurement uncertainties within a magnitude of 10 cm. This is further complemented by the significant reduction in data volume and object detection complexity, thus facilitating its deployment, primarily due to the reduced complexity of initial calibration, positioning, and deployment compared with three-dimensional segmentation algorithms.

Characterizing Satellite Geometry via Accelerated 3D Gaussian Splatting

The accelerating deployment of spacecraft in orbit has generated interest in on-orbit servicing (OOS), inspection of spacecraft, and active debris removal (ADR). Such missions require precise rendezvous and proximity operations in the vicinity of non-cooperative, possibly unknown, resident space objects. Safety concerns with manned missions and lag times with ground-based control necessitate complete autonomy. This requires robust characterization of the target’s geometry. In this article, we present an approach for mapping geometries of satellites on orbit based on 3D Gaussian splatting that can run on computing resources available on current spaceflight hardware. We demonstrate model training and 3D rendering performance on a hardware-in-the-loop satellite mock-up under several realistic lighting and motion conditions. Our model is shown to be capable of training on-board and rendering higher quality novel views of an unknown satellite nearly 2 orders of magnitude faster than previous NeRF-based algorithms. Such on-board capabilities are critical to enable downstream machine intelligence tasks necessary for autonomous guidance, navigation, and control tasks.

Simplified Maneuvering Strategies for Rendezvous in Near-Circular Earth Orbits

The development of autonomous guidance control and navigation systems for spacecraft would greatly benefit applications such as debris removals or on-orbit servicing, where human intervention is not practical. Within this context, inspired by Autonomous Vision Approach Navigation and Target Identification (AVANTI) demonstration, this work presents new guidance algorithms for rendezvous and proximity operations missions. Analytical laws are adopted and preferred over numerical methods, and mean relative orbital elements are chosen as state variables. Application times, magnitudes and directions of impulsive controls are sought to minimize propellant consumption for the planar reconfiguration of the relative motion between a passive target spacecraft and an active chaser one. In addition, simple and effective algorithms to evaluate the benefit of combining in-plane and out-of-plane maneuvers are introduced to deal with 3D problems. The proposed new strategies focus on maneuvers with a dominant change in the relative mean longitude (rarely addressed in the literature), but they can also deal with transfers where other relative orbital elements exhibit the most significant variations. A comprehensive parametric analysis compares the proposed new strategies with those employed in AVANTI and with the global optimum, numerically found for each test case. Results are similar to the AVANTI solutions when variations of the relative eccentricity vector dominate. Instead, in scenarios requiring predominant changes in the relative mean longitude, the required ΔV exhibits a 49.88% reduction (on average) when compared to the original methods. In all the test cases, the proposed solutions are within 3.5% of the global optimum in terms of ΔV. The practical accuracy of the presented guidance algorithms is also tested with numerical integration of equations of motion with J2 perturbation.

Three-Dimensional Autonomous Guidance for Enclosing a Stationary Target Within Arbitrary Smooth Geometrical Shapes

Three-Dimensional Autonomous Guidance for Enclosing a Stationary Target Within Arbitrary Smooth Geometrical Shapes

IoV and blockchain-enabled driving guidance strategy in complex traffic environment

Diversified traffic participants and complex traffic environment (e.g., roadblocks or road damage exist) challenge the decision-making accuracy of a single connected and autonomous vehicle (CAV) due to its limited sensing and computing capabilities. Using Internet of Vehicles (IoV) to share driving rules between CAVs can break limitations of a single CAV, but at the same time may cause privacy and safety issues. To tackle this problem, this paper proposes to combine IoV and blockchain technologies to form an efficient and accurate autonomous guidance strategy. Specifically, we first use reinforcement learning for driving decision learning, and give the corresponding driving rule extraction method. Then, an architecture combining IoV and blockchain is designed to ensure secure driving rule sharing. Finally, the shared rules will form an effective autonomous driving guidance strategy through driving rules selection and action selection. Extensive simulation proves that the proposed strategy performs well in complex traffic environment, mainly in terms of accuracy, safety, and robustness.

Vision-Based Autonomous Steering of a Miniature Eversion Growing Robot

Vision-Based Autonomous Steering of a Miniature Eversion Growing Robot

Advancing Autonomous Navigation: Deep Learning Techniques for Self-Driving Cars

Abstract: The fast development of self-driving car technology has brought forth radical breakthroughs in the world of transportation and movement. This study paper looks into the paradigm shift allowed by deep learning methods in getting stable and reliable autonomous steering for self-driving cars. Leveraging the power of artificial intelligence, particularly deep learning models, this study presents a comprehensive exploration of the multifaceted aspects involved in developing a self-driving car system that can adeptly perceive its environment, make intelligent decisions, and ensure passenger and pedestrian safety. The paper first reviews the basic concepts of self-driving cars, getting into the challenges associated with real-time awareness, scene understanding, and decision-making in complex and dynamic settings. It then shows the important role that deep learning plays in handling these challenges by studying various neural network designs designed for different tasks, including picture recognition, object detection, semantic segmentation, and path planning. The study encompasses convolutional neural networks (CNNs) for picture analysis.

A Study on the Rapid Detection of Steering Markers in Orchard Management Robots Based on Improved YOLOv7

In order to guide the orchard management robot to realize autonomous steering in the row ends of a complex orchard environment, this paper proposes setting up steering markers in the form of fruit trees at the end of the orchard rows and realizing the rapid detection of the steering markers of the orchard management robot through the fast and accurate recognition and classification of different steering markers. First, a high-precision YOLOv7 model is used, and the depthwise separable convolution (DSC) is used instead of the 3 × 3 ordinary convolution, which improves the speed of model detection; at the same time, in order to avoid a decline in detection accuracy, the Convolutional Block Attention Module (CBAM) is added to the model, and the Focal loss function is introduced to improve the model’s attention to the imbalanced samples. Second, a binocular camera is used to quickly detect the steering markers, obtain the position information of the robot to the steering markers, and determine the starting point position of the robot’s autonomous steering based on the position information. Our experiments show that the average detection accuracy of the improved YOLOv7 model reaches 96.85%, the detection time of a single image reaches 15.47 ms, and the mean value of the localization error is 0.046 m. Comparing with the YOLOv4, YOLOv4-tiny, YOLOv5-s, and YOLOv7 models, the improved YOLOv7 model outperforms the other models in terms of combined detection time and detection accuracy. Therefore, the model proposed in this paper can quickly and accurately perform steering marker detection and steering start point localization, avoiding problems such as steering errors and untimely steering, shortening the working time and improving the working efficiency. This model also provides a reference and technical support for research on robot autonomous steering in other scenarios.