Kaige Shi

Title: Kaige Shi and Human-Safe Robot Manipulation.

Introduction

Kaige Shi is an inventor associated with Nanyang Technological University in Singapore. His patent work addresses how a robot manipulator can operate safely in a workspace shared with people while carrying out a task.

Latest Patents

The supplied information lists one patent for Shi: “Method and system for controlling a robot manipulator for operating in a shared workspace with human(s).” The invention describes a control method that combines safety requirements with task-related trajectory constraints.

The method determines a safety control set for selected parts of the robot and nearby people, using a safety control function. It also determines hard- and soft-constraint control sets for trajectory tracking in the robot component’s degrees of freedom. An optimization step then uses all three sets to select the robot’s control input.

For safety, the method uses a control model expressed in control-affine form and evaluates pairs of robot and human parts. For each pair, it applies a safety-distance function corresponding to a control barrier function, together with a sliding manifold configured according to that function and its relative degree with respect to the robot’s control input. The approach is designed to account for both safe separation and the demands of the assigned task.

Career Highlights

The available information identifies Shi’s affiliation with Nanyang Technological University and records 1 patent. His listed invention focuses on control strategies for collaborative robotics, where robots and people share a working environment.

Collaborations

Guoqiang Hu is identified as Shi’s coworker. The supplied information does not specify the nature of their collaboration or whether they jointly developed the listed patent.

Conclusion

Kaige Shi’s listed patent presents a framework for coordinating robot safety and task performance in shared human-robot workspaces. By combining safety controls, trajectory constraints, and control-input optimization, the invention addresses a central challenge in collaborative robotics.