Font Size: a A A

Fast Optimization Of Asteroid Soft Landing Trajectory Based On Deep Neural Network And Shape Function

Posted on:2022-01-31Degree:MasterType:Thesis
Country:ChinaCandidate:W Y FengFull Text:PDF
GTID:2492306572955799Subject:Aeronautical and Astronautical Science and Technology
Abstract/Summary:PDF Full Text Request
With the continuous development of human deep space exploration technology,the exploration of planets in the solar system,especially asteroids,has become a focus in the field of deep space exploration.The detection methods of asteroids are more and more advanced,and the understanding of asteroids is more and more in-depth.However,due to the irregular shape,uneven mass distribution and unstable spin state of asteroids,the irregular gravitational field of asteroids also brings new challenges to the further exploration of asteroids.Therefore,this paper studies the fast optimization of asteroid landing trajectory in irregular gravitational field.The following is the main research work of this paper:Firstly,the dynamics of the landing phase near the asteroid is modeled,and the coordinate systems involved in the landing phase are defined.The traditional polyhedron method is used to model and analyze the gravitational acceleration near the asteroid 433 Eros,and a more accurate irregular gravitational field model is obtained,The orbit dynamics equation of asteroid landing phase is established.In addition,the polyhedral gravitational field model also provides a data set source for the later neural network model of gravitational field near asteroids.Secondly,the deep neural network is used to model the gravitational field near the asteroid.By building the deep neural network and taking the data set obtained from the polyhedral gravitational field model as the training sample of the neural network model,a more accurate neural network model of the gravitational field near the asteroid 433 Eros is obtained,The calculation speed of the neural network model is greatly improved,which lays the foundation for the rapid optimization of the subsequent landing trajectory.Finally,the orbit dynamics parameters of the landing phase are expanded in the form of Bezier shape function,and the corresponding constraints are derived considering the special characteristics of the asteroid,such as the irregular surface.Taking the landing exploration of 433 Eros asteroid as an example,the fast trajectory optimization of 433 Eros asteroid in the irregular gravitational field is studied by using Bezier shape function method,The applicability and efficiency of Bezier shape function method to trajectory optimization in non spherical gravitational field is verified.
Keywords/Search Tags:asteroid soft landing, deep neural network, Bezier shape function, trajectory optimization
PDF Full Text Request
Related items