Shumpei Tokuda

Papers

Total Citations

H-Index

About

Shumpei Tokuda is a robotics researcher focused on advancing task planning and manipulation for autonomous systems, particularly dual-arm robots. His work centers on integrating formal methods, such as Linear Temporal Logic (LTL), with motion planning to enable robots to autonomously generate and execute complex, multi-step tasks like pick-and-place operations. Tokuda’s major contribution is the development of fast, flexible planning algorithms that bridge high-level task specifications and low-level control, allowing robots to reason about object handling and sequencing without extensive manual programming. His most cited paper, “Fast LTL-Based Flexible Planning for Dual-Arm Manipulation” (2020, 19 citations), demonstrates a method for automatically generating actions from simple definitions, addressing the growing need for robotic labor in industrial settings. In “Convex Approximation for LTL-based Planning” (2021, 6 citations), he further refines this approach for nonlinear dynamical systems, enhancing computational efficiency. Tokuda also explores hierarchical task decomposition in “Generating New Lower Abstract Task Operator using Grid-TLI” (2020, 2 citations), proposing ways to subdivide tasks for more efficient teaching and execution. His work is notable for making formal logic-based planning practical for real-world robotics, with potential applications in manufacturing and service automation.