Search for: All records

Movement in compliant mechanisms is achieved, at least in part, via deformable flexible members, rather than using articulating joints. These flexible members are traditionally modeled using finite element analysis (FEA)-based models. In this article, an alternative strategy for modeling compliant cantilever beams is developed with the objectives of reducing computational expense and providing accuracy with respect to design optimization solutions. The method involves approximating the response of a flexible beam with an n-link/m-joint pseudo-rigid-body dynamic model (PRBDM). Traditionally, static pseudo-rigid-body models (PRBMs) have shown an approximation of compliant elements using two or three revolute joints (2R/3R-PRBM). In this study, a more general nR-PRBDM model is developed. The first n resonant frequencies of the PRBDM are matched to exact or FEA solutions to approximate the response of the compliant system and compared with existing PRBMs. PRBDMs can be used for co-design studies of flexible structural members and are capable of modeling large deflections of compliant elements. We demonstrate PRBDMs that show dynamically accurate response for a random geometry cantilever beam by matching the steady-state and frequency response, with dynamical response accuracies up to 10% using a 5R-PRBDM. 

Reinforcement learning (RL) has recently shown promise in solving difficult numerical problems and has discovered non-intuitive solutions to existing problems. This study investigates the ability of a general RL agent to find an optimal control strategy for spacecraft attitude control problems. Two main types of Attitude Control Systems (ACS) are presented. First, the general ACS problem with full actuation is considered, but with saturation constraints on the applied torques, representing thruster-based ACSs. Second, an attitude control problem with reaction wheel based ACS is considered, which has more constraints on control authority. The agent is trained using the Proximal Policy Optimization (PPO) RL method to obtain an attitude control policy. To ensure robustness, the inertia of the satellite is unknown to the control agent and is randomized for each simulation. To achieve efficient learning, the agent is trained using curriculum learning. We compare the RL based controller to a QRF (quaternion rate feedback) attitude controller, a well-established state feedback control strategy. We investigate the nominal performance and robustness with respect to uncertainty in system dynamics. Our RL based attitude control agent adapts to any spacecraft mass without needing to re-train. In the range of 0.1 to 100,000 kg, our agent achieves 2% better performance to a QRF controller tuned for the same mass range, and similar performance to the QRF controller tuned specifically for a given mass. The performance of the trained RL agent for the reaction wheel based ACS achieved 10 higher better reward then that of a tuned QRF controller 

The Engineering Systems Design Lab (ESDL) at the University of Illinois introduced Strain-Actuated Solar Arrays (SASAs) as a solution for precise satellite Attitude Control System (ACSs). SASA is designed to provide active mechanical vibration (jitter) cancellation, as well as small slew maneuver capabilities to hold a pose for short time periods. Current SASA implementations utilize piezoelectric distributed actuators to strain deployable structures, and the resulting momentum transfer rotates the spacecraft bus. A core disadvantage, however, is small strain and slew capability. Initial SASA systems could help improve pointing accuracy, but must be coupled with another ACS technology to produce large reorientations. A novel extension of the original SASA system is presented here that overcomes the small-displacement limitation, enabling use of SASA as a sole ACS for some missions, or in conjunction with other ACSs. This extension, known as Multifunctional Structures for Attitude Control (MSAC), can produce arbitrarily-large rotations, and has the potential to scale to large spacecraft. The system utilizes existing flexible deployable structures (such as solar arrays or radiators) as multifunctional devices. This multi-role use of solar panels extends their utility at a low mass penalty, while increasing reliability of the spacecraft ACS.