SBIR-STTR Award

The proposed innovation uses the powerful function approximation capabilities of neural networks (NNs) to enable real-time trajectory correction for spacecraft with electric propulsion (EP). We use a NN to “learn” the relationship between state and control near a reference trajectory, then use the NN to optimally follow the reference path in the presence of uncertainty. This innovation applies recent advancements from the field of artificial intelligence to spacecraft guidance and control. NN technology for EP automation has two key

Benefits:
1) the accuracy and optimality of running an onboard sophisticated program, and 2) a low computational requirement similar to legacy linear control architectures. The innovation enables increased spacecraft autonomy and makes spacecraft robust to large errors or large changes in target trajectory. Legacy onboard control algorithms are incapable of maintaining a spacecraft in highly sensitive orbits such as Earth-Moon libration point orbits (including NRHOs). Spacecraft with EP systems currently rely on frequent ground contacts in order to get updated thrust instructions. The proposed innovation will enable spacecraft with EP to autonomously follow a nominal path or rendezvous with another spacecraft in sensitive regimes, without significant onboard computation. The innovation benefits a wide variety of NASA projects, particularly the Lunar Orbiting Platform - Gateway, which will operate in this sensitive regime with an electric propulsion based Power and Propulsion Element. Example applications and benefits of the proposed innovation include: reducing operational costs for constellations of spacecraft, enabling EP spacecraft to perform transfers which are too sensitive for ground-in-the-loop control, autonomous stationkeeping in sensitive orbits, and ground-based Monte Carlo analyses of sensitivity and/or contingency cases. Potential NASA Applications (Limit 1500 characters, approximately 150 words) This technology will enable spacecraft with EP to autonomously follow a nominal path or rendezvous with another spacecraft in sensitive regimes, without significant onboard computation and without frequent ground support. The innovation benefits a wide variety of NASA projects, particularly the Lunar Orbital Platform-Gateway, which will operate in this sensitive regime with an EP based Power and Propulsion Element. Other NASA missions with EP will also benefit from reduced fuel costs and reduced operations costs. Potential Non-NASA Applications (Limit 1500 characters, approximately 150 words) Commercial and other users of the technology will be similarly aided by new degrees of autonomy and precision flying. The innovation reduces the need for continuous real-time support, reducing costs and helping ensure high levels of service availability. As the industry transitions to large constellations, autonomy is essential to minimize human operator costs and enable new business models.

The proposed innovation, neural networks (NNs) for electric propulsion (EP) — NNEP, leverages the fundamental principles of optimal control (OC) and a rich field of recent advancements in the area of artificial intelligence to automate spacecraft maneuver correction, resulting in improved spacecraft maneuver accuracy, lowered operations complexity, and cost savings. NNs are used as function approximators, learning the complex relationship between spacecraft state and the costates defining the OC to return to a reference trajectory. The NNEP technology builds on the variety of technologies that exist for onboard navigation (such as GPS, CAPS, or OpNav). Once the spacecraft has generated a state estimate, it evaluates a pre-trained NN to find the corresponding costates (non-physical terms created in the process of solving an OC problem). The NNEP technology maps state errors to costates because the costates are always smoothly-varying, even for non-smooth OC. Within seconds of the nav update, the spacecraft autonomously determines the control required for the next several days and checks for constraint violations. The NNEP technology consists of both a novel application of NNs to relevant astrodynamics problems and an architecture for implementing this technology in real operational environments and in flight software (FSW). A proof of concept of the technology was developed during Phase I, including demonstration of the building blocks for a FSW implementation. Phase II funding will mature the technology further and result in a prototype FSW implementation running on representative space hardware. Many research groups are now investigating the use of NNs to automate spacecraft trajectory corrections. NNEP combines Advanced Space’s practical institutional experience of mission design, navigation, and operations for a wide variety of cutting-edge space missions with the powerful theoretical advantages of NNs and OC. Potential NASA Applications (Limit 1500 characters, approximately 150 words) The proposed innovation has wide applicability to NASA space missions. Some examples of where the autonomous trajectory correction capabilities would benefit operations include: the lunar Gateway or other elements of the Artemis program, including correction maneuvers en route to the NRHO, stationkeeping in the NRHO, and correction maneuvers for transfers between cislunar orbits; interplanetary trajectory correction; and constellation maintenance. All of these can be automated with either chemical or electric propulsion. Potential Non-NASA Applications (Limit 1500 characters, approximately 150 words) The proposed innovation has wide applicability to non-NASA space missions. Some examples of where the autonomous trajectory correction capabilities would benefit operations include: GEO stationkeeping; LEO constellation stationkeeping; commercial lunar landers; and electric propulsion (EP) transfers between GTO and GEO. All of these can be automated with either chemical or electric propulsion.