ZJUNlict Extended Team Description Paper 2025

TL;DR

ZJUNlict addresses robust autonomous play in RoboCup Small Size League by integrating an IMU into the v2023 robot and optimizing software for real-time decision-making. Hardware work centers on dribbler dynamics, high-frequency feedback, and IMU-based control, while software advances deliver improved ball pursuit and possession prediction, a redesigned strategy module, and a CUDA-driven decision pipeline. Key contributions include a simplified dribble model with simulation validation, high-resolution dribbler feedback via infrared sensing and LSTM prediction, and an IMU-enabled control loop with calibrated yaw alignment, all contributing to improved stability and responsiveness under high-tempo game dynamics. Collectively, these developments enhance on-field performance by enabling faster, more accurate posture control, predictive ball handling, and coordinated multi-robot decisions.

Abstract

This paper presents the ZJUNlict team's work over the past year, covering both hardware and software advancements. In the hardware domain, the integration of an IMU into the v2023 robot was completed to enhance posture accuracy and angular velocity planning. On the software side, key modules were optimized, including the strategy and CUDA modules, with significant improvements in decision making efficiency, ball pursuit prediction, and ball possession prediction to adapt to high-tempo game dynamics.

Figures (9)

• Figure 1: Simplified dribble system model
• Figure 2: Simulation results of the position of m with different M
• Figure 3: Dribblers of different joint positions
• Figure 4: Imu data collected when dribbling
• Figure 5: Comparison of Y-direction offset along the trackline with angular velocity planning using IMU versus vision-based planning.