Research
My work involves development of algorithms for autonomous motion planning and control of resource constrained robotic systems under uncertainty with application to spacecraft guidance and control. My research leverages tools from probability theory, nonlinear control theory, and convex optimization.
The above picture summarizes the different aspects involved in design and testing for autonomy. Implementing autonomy algorithms for spacecraft guidance and control poses additional challenges in terms of uncertain environment and computational power. Ongoing research handles known unknowns and unknown unknowns in motion planning and nonlinear control using techniques from machine learning and stochastic optimal control.
Research Projects
We present the Generalized Polynomial Chaos-based Sequential Convex Programming method to compute a sub-optimal solution for a continuous-time chance-constrained stochastic nonlinear optimal control problem (SNOC) problem. The approach enables motion planning and control of robotic systems under uncertainty. The proposed method involves two steps. The first step is to derive a deterministic nonlinear optimal control problem (DNOC) with convex constraints that are surrogate to the SNOC by using gPC expansion and the distributionally-robust convex subset of the chance constraints. The second step is to solve the DNOC problem using sequential convex programming (SCP) for trajectory generation and control.
We present an architecture for inspection or mapping of a target spacecraft, referred to as chief, in a low Earth orbit using multiple spacecraft, referred to as deputies (or) observers, in stable Passive Relative Orbits (PROs). We use an information gain approach to directly consider the trade-off between gathered data and fuel/energy cost. The four components of our architecture are: 1) information estimation, 2) state estimation, 3) motion planning for relative orbit initialization and reconfiguration, and 4) relative orbit control.
Learning-based control algorithms require data collection with abundant supervision for training. Safe exploration algorithms ensure the safety of this data collection process even when only partial knowledge is available. We present a new approach for optimal motion planning with safe exploration that integrates chance-constrained stochastic optimal control with dynamics learning and feedback control. We derive an iterative convex optimization algorithm that solves an Information-costStochastic Nonlinear Optimal Control problem (Info-SNOC). The optimization objective encodes control cost for performance and exploration cost for learning, and the safety is incorporated as distributionally robust chance constraints.
This paper presents a new six-degree-of-freedom robotic spacecraft simulator, the Multi-Spacecraft Testbed for Autonomy Research (M-STAR), for testing formation guidance, relative navigation, and control algorithms. The simulator dynamics are governed by five degrees of frictionless translational and rotational air-bearing motion and one degree of kinematic motion in the gravity direction with flight-like actuators, in a 1-g environment. M-STAR is designed to be modular and accommodates 3-DOF, 4-DOF, 5-DOF, and 6-DOF operation with minimal mechanical modifications. The simulator is modeled as a 3-D pendulum on a floating platform with sixteen thrusters and four reaction wheels as onboard actuators. Based on this plant model, a nonlinear hierarchical control law is proposed for position and attitude trajectory tracking.
This paper presents a novel control architecture and algorithm for precision attitude control of a one-degree-of-freedom dynamic model of a spacecraft using Strain Actuated Solar Arrays.
In this project, we design, implement and validate autonomy algorithms for autonomous on-orbit construction.