Characterisation of 6DOF natural and controlled relative dynamics in cislunar space

Novel Learning for Control of Nonlinear Spacecraft Dynamics

“…The program utilizes the ode45 solver to simulate the response of the overall system combining the controller and the controlled system. (6); Jt = [p.J(1,1);p.J(1,2);p.J(1,3);p.J(2,2);p.J(2,3);p.J(3,3)]; % time tfinal = 5; deltat = 0.001; t = 0:deltat:tfinal;% for evaluating solution % solve the ODE z0 = [1 0 0 0 0 0 0 0.08 0.08 0.04 1.06 0.21 0.31 1.90 0.41 1.15]'; options = odeset('absTol',1e-10,'relTol',1e-10); % The simulation for the general version of modified learning method [t_dai, z_dai] = ode45(@(t,z)deterministic_artificial_intelligence_modified_general(t,z,p), t, z0, options); % The simulation for the specific version of modified learning method [t_dmd, z_dmd] = ode45(@(t,z)deterministi_artificia_intelligence_modified_specific(t,z,p), t, z0, options); % The simulation for the original version of learning method [t_dor, z_dor] = ode45(@(t,z)deterministic_artificial_intelligence_original(t,z,p), t, [z0;0;0;0], options); %% Plot parameter estimations and state trajectories figure() plot(t_dai, z_dai(:,11),t_dai, z_dai(:, 14) 2) th(3); th(2) th( 4) th(5); th(3) th( 5) th( 6)]; % generate trajectory [sdwd,sddwd] = traj_gen(t,wd,p); % generate feed forward control torque tau = Jh*sdwd + cross(wd', Jh*wd)'; % update the dynamic of the system dw = p.J\(tau-cross(w', p.J*w)'); dq = 0.5*quatmultiply([0 w'],q'); % update the parameter estimation ddw = [0;0;0]; Pd = p.P(wd(1),wd(2),wd(3),sdwd(1),sdwd(2),sdwd(3),sddwd(1),sddwd(2),sddwd(3)); P = p.P(w(1),w(2),w(3),dw(1),dw(2),dw(3),0,0,0); dP = p.dP(w(1),w(2),w(3),dw(1),dw(2),dw(3),ddw(1),ddw(2),ddw…”

“…al, [5] proposed a chattering attenuation sliding mode control utilizing the eigen structure of the linearized flow around a libration point of the Earth-Moon circular restricted three-body problem, and this novel article serves as a reminder of the prevalence of linearization when dealing with multiple, coupled nonlinear equations. In 2021, Colombia presented a guidance, navigation and control framework for 6 degrees of freedom (6DOF) coupled Cislunar rendezvous and docking, and the article highlighted the importance of dealing with full, coupled translational-rotational dynamics of multi-body (i.e., highly flexible) dynamics seeking guaranteed coupled-state estimation [6]. Immediately that same year [7], new techniques for highly flexible multi-body space robotics were proposed as a competing narrative to the just-proposed "whiplash compensation" of flexible space robotics [8].…”

Guidance, navigation and control for 6DOF rendezvous in Cislunar multi-body environment

“…Alternative control parameterizations have been investigated in previous literature works [24,28,40], but the one discussed in this section guarantees a good compromise between robustness and fast convergence of the guidance and control algorithm for the considered example applications. The authors suggest leaving enough freedom in setting the control parametrization for alternative problems.…”

Optical navigation for Lunar landing based on Convolutional Neural Network crater detector

“…Lunar environment and Artificial Intelligence are becoming increasingly attractive to the Space research community, due to the latest long-term plans of Space Agencies. On one hand, the activities linked to the Lunar Gateway have renovated the deep interest in the mentioned environment [1,2,3,4].…”