Development of machine learning (ML) applications is hard. Producing successful applications requires, among others, being deeply familiar with a variety of complex and quickly evolving application programming interfa...
详细信息
ISBN:
(纸本)9781450375078
Development of machine learning (ML) applications is hard. Producing successful applications requires, among others, being deeply familiar with a variety of complex and quickly evolving application programming interfaces (APIs). It is therefore critical to understand what prevents developers from learning these APIs, using them properly at development time, and understanding what went wrong when it comes to debugging. We look at the (lack of) guidance that currently used development environments and ML APIs provide to developers of ML applications, contrast these with software engineering best practices, and identify gaps in the current state of the art. We show that current ML tools fall short of fulfilling some basic software engineering gold standards and point out ways in which software engineering concepts, tools and techniques need to be extended and adapted to match the special needs of ML application development. Our findings point out ample opportunities for research on ML-specific software engineering.
This paper studies the problem of tracking a ground target for a fixed-wing unmanned aerial vehicle(UAV) based on the proposed guidance law. The algorithm ensures that a UAV continuously overflies the target whether i...
详细信息
This paper studies the problem of tracking a ground target for a fixed-wing unmanned aerial vehicle(UAV) based on the proposed guidance law. The algorithm ensures that a UAV continuously overflies the target whether it is fixed or moving. The requirements of the UAV flight constraints such as bounded airspeed and acceleration are considered. A Lyapunov function is constructed to prove the stability of the proposed guidance law,and parameter design criteria have been developed. Considering the fixed and moving ground targets, numerical simulations are performed to verify the feasibility and benefits of the proposed guidance algorithm.
The threat of drones in sensitive airspace is a growing issue. Drone detection currently focuses on security threats of undesired surveillance. In this thesis, we seek to determine whether a supervisory controller app...
详细信息
The threat of drones in sensitive airspace is a growing issue. Drone detection currently focuses on security threats of undesired surveillance. In this thesis, we seek to determine whether a supervisory controller applied to a quadrotor drone will suffice as a feasible option for an autonomous drone. This is achieved by applying a proportional and derivative control law, programmed within MATLAB, to a pre-built simulation model and implemented on the Parrot Mambo Drone for experimental flights. The analysis is accomplished by comparing the performance and accuracy of the simulated trial flight, an experimental flight on the Parrot Mambo Drone with no altitude change, and an experimental flight with adjustments in all six degrees of freedom on the Parrot Mambo Drone. The results show that with a supervisory controller applied to a quadrotor drone, the drone can perform the desired tasks autonomously at a higher standard than without a supervisory controller applied. The proportional and derivative controls implemented on the Parrot Mabo Drone for the experimental flights have the best performance of the control law investigated.
To achieve the goal of collecting lunar samples and return to the Earth for the Chang'E-5 spacecraft,the lander and ascender module(LAM)of the Chang'E-5 spacecraft successfully landed on the lunar surface on 1...
详细信息
To achieve the goal of collecting lunar samples and return to the Earth for the Chang'E-5 spacecraft,the lander and ascender module(LAM)of the Chang'E-5 spacecraft successfully landed on the lunar surface on 1 Dec.,*** guidance,navigation,and control(GNC)system is one of the critical systems to perform this *** GNC system of previous missions,Chang'E-3 and Chang'E-4,provides the baseline design for the Chang'E-5 LAM,and the new characteristics of the LAM,like larger mass and liquid sloshing,also bring new challenges for the GNC *** GNC design for the descent and landing is presented in this *** guidance methods implemented in the powered descent are presented in detail for each *** consumption and hazard avoidance should be particularly considered in the design.A reconfigurable attitude control is adopted which consists of the quaternion partition control,phase and gain stabilization filter,and dual *** controller could provide fast attitude maneuver and better system *** the navigation,an intelligent heterogeneous sensor data fusion method is presented,and it is applied for the inertial measurement unit and velocimeter ***,the flight results of the LAM are *** sensors were able to provide valid measurement data during descent,and the thrusters and the main engine operated well as ***,a successful soft lunar landing was achieved by the LAM.
To perform an accurate fire-and-forget attack of seeker-less missile with the consideration of the localization error, a new cooperative guidance strategy of multi-missiles is proposed in this work. Firstly, an estima...
详细信息
To perform an accurate fire-and-forget attack of seeker-less missile with the consideration of the localization error, a new cooperative guidance strategy of multi-missiles is proposed in this work. Firstly, an estimated position of the seeker-less missile is carefully selected, based on which the slant angle of the line between the estimated and real positions of the seeker-less missile is approximately derived inspired by the localization theory of GPS(Global Position System).Then, respectively based on the guidance law considering the impact angle and time constraints, the consensus theory and the finite-time convergence approach, three cooperative guidance laws employing the previously obtained slant angle are derived for intensive attack to address either a stationary or maneuvering target. Simulation results well demonstrate the effectiveness and advantages of the proposed cooperative guidance strategy.
A smart homing guidance strategy with control saturation against a target-defender team is derived. It is noteworthy that a cooperative strategy of the target-defender team is applied,which has been proved more challe...
详细信息
A smart homing guidance strategy with control saturation against a target-defender team is derived. It is noteworthy that a cooperative strategy of the target-defender team is applied,which has been proved more challenging for the homing *** defender missile is launched by the target and guided by a cooperative augmented proportional navigation(APN). At the same time, the target performs a one-switch maneuver to cooperate and minimize the defender's acceleration requirement. The problem is analyzed for arbitrary-order linear dynamics of the agents in the linearized form but validated by the mathematical simulations by using nonlinear kinematics. The perfect information of three agents' states is assumed. Then, a method to deal with the target-defender team is proposed. It contains a combined performance index penalizing the miss distance relative to the target and energy consumption in the whole duration. Besides, the specific miss distance related to the defender is regarded as an inequality constraint. An analytical solution for the smart guidance strategy against the APN guided defender is derived. Meanwhile, the control saturations are introduced to get more realistic and reasonable insights to this practical target-missile-defender problem. A simple but effective iterative searching technique is proposed to determine the saturation time points. The solution provides an optimal homing strategy to evade the defender with a specific miss distance and intercept the target with the minimum miss distance in the minimum energy manner. Nonlinear two-dimensional simulation results are used to validate the theoretical analysis. By comparison with the optimal differential game guidance(ODGG) and the combined minimum effort guidance(CMEG), the superiority of this smart guidance strategy is concluded.
In this paper, it is described as an uncertain model under an integrated style of control and guidance. Then, a nonlinear observer is designed for it. For view, we have some results of its simulation. It shows that th...
详细信息
When GPS is not available, autonomous vehicles can obtain navigation information by observing a source with a camera or an acoustic system mounted on the frame of the vehicle. This information can be used in the navig...
详细信息
When GPS is not available, autonomous vehicles can obtain navigation information by observing a source with a camera or an acoustic system mounted on the frame of the vehicle. This information can be used in the navigation and to overcome the lack of other sources of positioning. However, these systems often have a limited range and angular FOV. Due to this restriction, motion along some paths will make it impossible to obtain the necessary navigation information as the source is no longer in the vehicle's FOV. This paper proposes both a path planning approach and a guidance control law that allows the vehicle to preserve landmarks that are critical for the navigation inside the FOV while at the same time converging to the proposed path. The suggested approach is experimentally validated by implementation in an autonomous underwater vehicle.
The guidance of an automated vehicle requires a software-module, which plans a comfortable and collision-free future motion. Furthermore, it must be able to lead the vehicle into a given planning goal. This requires t...
详细信息
ISBN:
(纸本)9783800754878
The guidance of an automated vehicle requires a software-module, which plans a comfortable and collision-free future motion. Furthermore, it must be able to lead the vehicle into a given planning goal. This requires the design of a trajectory planning method including a suitable objective function under consideration of vehicle dynamics and environmental constraints. In this contribution, a trajectory planning algorithm is proposed, which enables the execution of different maneuvers for on-road driving. Restrictions of the state- and input space are included as hard constraints.
With the development of aerospace industry, the guidance system of an entry vehicle will be more robust, reliable and autonomous in the future. Based on fuzzy logic, the paper designs a predictor-corrector guidance la...
详细信息
With the development of aerospace industry, the guidance system of an entry vehicle will be more robust, reliable and autonomous in the future. Based on fuzzy logic, the paper designs a predictor-corrector guidance law. The trajectory prediction is realized by numerical integration. The correction system is based on two fuzzy controllers, which correct the longitudinal motion and lateral motion synergistically. The error of flight range is eliminated by correcting the magnitude of bank angle. The altitude error is eliminated by correcting the attack angle. The lateral error is eliminated by regulating the reversal time of bank angle. Comparing with the traditional corrector based on Newton-Raphson iteration, the method in this paper only needs a single trajectory prediction in one correction cycle, which is favorable for on-board calculation. Moreover, the longitudinal motion and lateral motion are synergistically corrected in the predictor-corrector, which makes the method more robust and flexible.
暂无评论