Takeshi Uchitane

The University of Osaka, Aichi Institute of Technology

Papers

Total Citations

H-Index

About

Takeshi Uchitane is a leading researcher in multi-agent robotics and simulation, with a primary focus on the RoboCup Soccer Simulation leagues. His work bridges evolutionary computation and robotics, most notably through pioneering methods for bipedal locomotion. In his highly cited 2011 paper, Uchitane applied evolution strategies to tune gait patterns for humanoid robots in the RoboCup 3D Soccer Simulation environment, combining nonlinear oscillators with PD controllers to generate stable, rhythmic walking. This approach, refined in his 2010 work, demonstrated how evolutionary algorithms could automatically optimize complex parameter sets—a significant step toward adaptive, autonomous locomotion. More recently, Uchitane has advanced game intelligence in the RoboCup 2D Simulation league by developing the concept of Expected Possession Value (EPV), a metric that enables agents to dynamically evaluate ball control probabilities in real time. His contributions have earned over 15 citations across his most-cited works, reflecting their impact on both robotics and simulation research. Through his iterative work on locomotion and strategic decision-making, Uchitane continues to shape how simulated agents learn and adapt in complex, competitive environments.