Research
--- Publications ---
Alphonsus Adu-Bredu*, Grant Gibson*, and Jessy W. Grizzle. "Exploring kinodynamic fabrics for reactive whole-body control of underactuated humanoid robots". In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Oluwami Dosunmu-Ogunbi, Aayushi Shrivastava, Grant Gibson, Jessy W Grizzle. "Stair Climbing using the Angular Momentum Linear Inverted Pendulum Model and Model Predictive Control". In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
Grant Gibson, Oluwami Dosunmu-Ogunbi, Yukai Gong, and Jessy Grizzle. "Terrain-adaptive, alip-based bipedal locomotion controller via model predictive control and virtual constraints". In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
*: Equal Contribution
--- Current Projects ---
Perception-Integrated, Terrain-Aware Humanoid Locomotion Controller
Summary: Implementation of my ALIP-MPC controller on the Digit robot using perception data for terrain parameter estimation.
Angular Momentum Based Running Controller
Summary: An extension and improvement upon Yukai Gong's preliminary results. This controller executes desired stance phase and flight phase trajectories based upon the predicted angular momentum at each impact.
--- Publication-related Projects ---
Exploring Kinodynamic Fabrics for Reactive Whole-Body Control of Underactuated Humanoid Robots (In Review)
Alphonsus Adu-Bredu*, Grant Gibson*, and Jessy W. Grizzle
*: Equal Contribution
Summary: We propose Kinodynamic Fabrics as an approach for the specification, solution and simultaneous execution of multiple motion tasks in real-time while being reactive to dynamism in the environment. Kinodynamic Fabrics allows for the specification of prioritized motion tasks as forced spectral semi-sprays and solves for desired robot joint accelerations at real-time frequencies. We evaluate the capabilities of Kinodynamic fabrics on and diverse physically-challenging whole-body control tasks with a bipedal humanoid robot both in simulation and in the real-world.
Stair Climbing using the Angular Momentum Linear Inverted Pendulum Model and Model Predictive Control (In Review)
Oluwami Dosunmu-Ogunbi, Aayushi Shrivastava, Grant Gibson, Jessy W Grizzle
Summary: This paper develops a controller that enables the Cassie robot to climb stairs in simulation. Novel contributions include the exploitation of a variation of the ALIP model that allows CoM height to vary within a step, and a novel combination of virtual constraint-based control and MPC to stabilize a stair-climbing gait.
Terrain-Adaptive, ALIP-Based Bipedal Locomotion Controller via Model Predictive Control and Virtual Constraints (IROS 2022)
--- Additional Projects ---
ROS 2 Workspace for Digit Robot Python SDK
Summary: ROS 2 workspace with packages to interface with the Python SDK for the Digit robot designed by Agility Robotics.
Repository found here
Angular Momentum Linear Inverted Pendulum (ALIP) Controllers: Point Mass & 5-Link Walker
Input-Output Nonlinear Model Predictive Control of Five-Link Robot for Stair Climbing
Summary: Preliminary work on achieving robust stair climbing for a planar five-link robot. Offline gait trajectories were generated using the Hybrid Zero Dynamics Framework. A novel control algorithm was developed by combining input-output linearization and nonlinear model predictive control.
Repository found here
Three-Link Hybrid Zero Dynamics Gait Design & Control
Summary: Hybrid Zero Dynamics Optimization and Simulation for a 3-link walker. Based upon the work presented in Feedback of Dynamic Bipedal Robot Locomotion by Eric Westervelt and Jessy W. Grizzle
Repository found here
Acrobot Control Algorithms
Summary: Implemention of various control strategies for the acrobot (double pendulum model). Fully-actuated and underactuated cases are explored for the pinned model. Passive walking and Passivity-based control methods are also implemented for the floating base model walking at an incline.
Repository found here