The present disclosure provides a method for controlling orbital trajectories of a plurality of spacecraft in a multi-spacecraft swarm. In one aspect, the method includes deploying a DRL agent including a plurality of trajectory control models to the multi-spacecraft swarm, the trajectory control models corresponding to swarm configurations of the multi-spacecraft swarm; determining a state vector of said plurality of spacecraft in the multi-spacecraft swarm; transmitting a collective command to the multi-spacecraft swarm, such that said plurality of spacecraft in the multi-spacecraft swarm are to be distributed in one of the swarm configurations; determining actions of said plurality of spacecraft based on the state vector and the collective command; and maneuvering the multi-spacecraft swarm in accordance with the actions.

A method for capturing and deorbiting space debris includes: providing a space debris capturing device; deploying the space debris capturing device in planetary orbit; determining, via an onboard global positioning system unit, the position and orbit velocity of the space debris capturing device; receiving an initial target set including a first database of space debris targets that are within range of the space debris capturing device; performing a first algorithm to convert the initial target set to an accessible target set including a second database of space debris targets that are within range of the space debris capturing device, the second database is smaller than the first database; performing a second algorithm to convert the accessible target set to a final target set including a third database of space debris targets to be captured by the space debris capturing device, the third database is smaller than the second database; transferring the space debris capturing device to a position within a capture range of a first space debris target of the third database; capturing the first space debris target via a capture mechanism of the space debris capturing device; jettisoning the capture mechanism and the first captured space debris target into a decaying orbit; repeating the transferring, capturing, and jettisoning steps for all but a final one of the remaining space debris targets of the third database; and positioning the space debris capturing device and the final captured space debris target into a decaying orbit.

A spacecraft including: an engine; a thrust vector control device controlling a thrust vector as a direction of a thrust acting on the spacecraft; and a main control device configured to acquire state quantities of the spacecraft in a powered descending in which the spacecraft is guided to a target point while the engine generates the thrust, and generate a throttling command by which combustion of the engine is controlled and an operation command by which the thrust vector control device is operated. The state quantities contain a first acceleration parameter and a second acceleration parameter. The first and second acceleration parameters are calculated as coefficients A and B obtained by fitting based on acceleration of the spacecraft previously detected, supposing the following equation is satisfied between a reciprocal number 1/a of the acceleration a of the spacecraft and time t:
1/a=−At+B  (1).

The invention is methods for landing rockets with precision using deep reinforcement learning for control. Embodiments of the invention are comprised of three steps. First, sensors collect data about the rocket's physical landing environment, passing information to rocket's database and processors. Second, the processors manipulate the information with a deep reinforcement learning program to produce instructions. Third, the instructions command the rocket's control system for optimal performance during landing.

In an example, a method for deorbiting a spacecraft is described. The method includes selecting a target landing site for deorbiting the spacecraft. The method includes determining a range target and a velocity target for reaching a predicted atmospheric entry location. The method includes determining a back-propagated orbit state estimate of the spacecraft. The method includes comparing the back-propagated orbit state estimate to a known orbit state of the spacecraft to determine that the back-propagated orbit state estimate has converged with the known orbit state. The method includes calculating based on determining that the back-propagated orbit state estimate has converged with the known orbit state, (a) an estimated time of ignition for a propulsion system of the spacecraft and (b) an estimated burn velocity vector of the propulsion system using the range target and the velocity target. The method includes performing a burn pulse by the propulsion system.

Systems and methods for describing, simulating and/or optimizing spaceborne systems and missions. Configurations for spaceborne systems are generated and validated based on simulation output.

A system detects a maneuver of at least one space object by receiving a first data set relating to orbital characteristics of at least one space debris object. The system trains a model, using the first data set, in order to model orbital behaviors of the at least one space debris object. The system then receives a second data set relating to orbital characteristics of the at least one space object, and detects a maneuver of the at least one space object using the trained model and the second data set.

A lunar orbiting satellite system executes orbit planning of assigning a function (positioning, communication, and flashing) to an artificial satellite (AS) depending on a relative position of the AS to the moon at a time when the moon and the AS are observed from an input point on the earth, and correcting the relative position, which changes in accordance with the moon revolution period. The system includes: a satellite orbit planner which assigns a function to each ASs forming an AS group flying around the moon depending on a relative position of each ASs to the moon at a time when the moon and ASs are observed from an input point on the earth, and set a target orbit according to the function; and a satellite controller which causes the each ASs to execute control based on the function to implement switching of the function.

For power-enhanced slew maneuvers, a method determines a power collection function for a satellite. The method determines a power cost function for the satellite. The method calculates a power enhanced slew maneuver based on the power collection function and the power cost function.

A Vehicle Based Independent Range System (VBIRS) (10) comprised of individual stacked chambered modules that function as a single integrated system that provides a self-contained space based range capability, and is comprised of a power module (12), an artificial intelligence/autonomous engagement/flight termination system module (20), a satellite data modem module system (30) and a navigation, communications and control module system (40), all of which interface with a VBIRS test and checkout system (52) and a weather data system (116). The artificial intelligence/autonomous engagement/flight termination system module (20) is comprised of an inherent artificial intelligence capability that envelopes and interchanges data with an autonomous engagement controller (22) that contains all missile/rocket autonomous cooperative engagement, destruct decision software and range safety algorithm parameters required for optimum mission planning. VBIRS employed aboard an aircraft or between any combination of launching systems allows that aircraft to launch a missile/rocket from any location on earth, whether the missile/rocket is singularly launched by itself or as a larger group of missiles/rockets launched in a salvo arrangement, while providing collaborative real-time targeting to occur directly between missiles/rockets in conjunction with other missile/rocket launch platforms or stand-alone mission control centers.