Reconstruction and Cognitive Learning Optimization of Unmanned Bicycle Motion Controller

doi:10.3901/JME.2025.07.301

Abstract

Abstract: Aiming at the unknown structure and parameters of the balance controller of unmanned bicycle and the lack of generalization ability of the control system, a cognitive learning system for reconstructing the unknown controller and improving the dynamic performance of the control system was proposed. The cognitive learning system is mainly composed of two parts: One is to use Gaussian process regression (GPR) theory to establish a probability model to represent the mapping relationship between the input and output of the controller, taking the lateral inclination Angle, lateral inclination speed, handlebar Angle and handlebar Angle speed as inputs and the control torque of the handlebar as outputs. Second, combining the reactive cognitive learning optimization strategy, the perception module, cognitive module and executive module are introduced, and the output of the established probability model is taken as the initial training value of the cognitive learning controller of the unmanned bicycle, and the control performance of the controller is iteratively optimized. The results of numerical simulation and prototype experiment show that the Gaussian process regression method can reconstruct the unknown controller well, and the cognitive learning algorithm can effectively improve the dynamic performance of the control system. The combination of the two enables the unmanned bicycle to achieve stable and balanced walking on a relatively flat cement road surface, and its lateral inclination Angle fluctuates within the range of [-0.042rad, 0.044rad]. Compared with the GPR controller, the optimized cognitive learning system reduces the lateral inclination Angle of the vehicle body and the amplitude of the handlebar Angle by about 55.7% and 57.1%, respectively. It also shows good generalization performance for the running scenarios of wet road, slope, lawn.

Key words: unmanned bicycle, balance exercise, controller reconfiguration, cognitive learning, Gaussian process regression

CLC Number:

TP242

HUANG Yonghua, LIANG Ziyan, ZHUANG Wei, YANG Haiyang. Reconstruction and Cognitive Learning Optimization of Unmanned Bicycle Motion Controller[J]. Journal of Mechanical Engineering, 2025, 61(7): 301-314.

References

[1] BEZNOS A V，FORMAl’SKY A M，GURFINKEL E V，et al. Control of autonomous motion of two-wheel bicycle with gyroscopic stabilisation[C]// Proceedings. 1998 IEEE International Conference on Robotics and Automation (Cat. No. 98CH36146). IEEE，1998，3：2670-2675.
[2] YETKIN H，KALOUCHE S，VERNIER M，et al. Gyroscopic stabilization of an unmanned bicycle[C]// American Control Conference (ACC)，2014. IEEE，2014：4549-4554.
[3] JIN H Z，YANG D C，LIU Z X，et al. A gyroscope-based inverted pendulum with application to posture stabilization of bicycle vehicle[C]// Proceedings of the 2015 IEEE Conference on Robotics and Biomimetics，Zhuhai，China. Zhuhai：IEEE，2015：2103-2108.
[4] SUEBSOMRAN A. Balancing control of bicycle robot[C]// Cyber Technology in Automation，Control，and Intelligent Systems (CYBER)，2012 IEEE International Conference on. IEEE，2012：69-73.
[5] KIM Y，KIM H，LEE J. Stable control of the bicycle robot on a curved path by using a reaction wheel[J]. Journal of Mechanical Science and Technology，2015，29(5)：2219-2226.
[6] KOOIJMAN J D G，MEIJAARD J P，PAPADOPOULOS J M，et al. A bicycle can be self-stable without gyroscopic or caster effects[J]. Science，2011，332(6027)：339-342.
[7] 郭磊，黄用华，廖启征，等. 自平衡自行车机器人的运动学分析[J]. 北京邮电大学学报，2011，34(6)：99-102. GUO Lei，HUANG Yonghua，LIAO Qizheng，et al. Kinemics analysis for a self-balancing bicycle robot[J]. Journal of Beijing University of Posts and Telecommunications，2011，34(6)：99-102.
[8] 李艳，王涵. 基于复合控制的自行车机器人平衡控制[J]. 现代电子技术，2019，42(2)：177-182. LI Yan，WANG Han. Balance control of bicycle robot based on compound control[J]. Modern Electronics Technique，2019，42(2)：177-182.
[9] 艾红，童璐，李成荣. 无机械辅助结构自行车机器人控制仿真及实现[J]. 信息与控制，2019，48(2)：187-193. AI Hong，TONG Lu，LI Chengrong. Control simulation and implementation of a bicycle robot without a mechanically assisted structure[J]. Information and Control，2019，48(2)：187-193.
[10] 庄未，苏晓，赵逸舟，等. 车把前倾角对无人驾驶自行车90°圆周运动的影响[J]. 机械工程学报，2021，57(2)：256-264. ZHUANG Wei，SU Xiao，ZHAO Yizhou，et al. Influence of the handle forward bar angle of riderless bicycle on the 90° circular motion[J]. Journal of Mechanical Engineering，2021，57(2)：256-264.
[11] HUANG Y H，LIAO Q Z，GUO L，et al. Simple realization of balanced motions under different speeds for a mechanical regulator-free bicycle robot[J]. Robotica，2014，72(9)：1-15.
[12] HUANG Y H，LIAO Q Z，GUO L，et al. Balanced motions realization for a mechanical regulators free and front-wheel drive bicycle robot under zero forward speed[J]. International Journal of Advanced Robotic Systems，2013，10(317)：1-9.
[13] BAQUERO-SUÁREZ M，CORTÉS-ROMERO J，ARCOS- LEGARDA J，et al. A robust two-stage active disturbance rejection control for the stabilization of a riderless bicycle[J]. Multibody System Dynamics，2019，45(1)：7-35.
[14] 黄用华，林英鹏，庄未，等. 无人自行车侧向平衡的全阶滑模控制器设计与分析[J]. 机械设计与研究，2022，38(2)：12-17. HUANG Yonghua，LIN Yingpeng，ZHUANG Wei，et al. Design and analysis of a full-order sliding mode lateral balance controller for an unmanned bicycle[J]. Machine Design & Research，2022，38(2)：12-17.
[15] KIM H，CHANG H. Model reference Gaussian process regression：data-driven state feedback controller[J]. IEEE Access，2023，11：134374-134381.
[16] MAHLER J，KRISHNAN S，LASKEY M，et al. Learning accurate kinematic control of cable-driven surgical robots using data cleaning and gaussian process regression[C]// 2014 IEEE international conference on automation science and engineering (CASE). IEEE，2014：532-539.
[17] MORIMOTO J，ATKESON C G，ENDO G，et al. Improving humanoid locomotive performance withlearnt approximated dynamics via Gaussian processes for regression[C]// Intelligent Robotsand Systems，2007. IROS 2007. IEEER SJ International Conference on. 2007：4234-4240.
[18] 陈友东，郭佳鑫，陶永. 基于高斯过程的机器人自适应抓取策略[J]. 北京航空航天大学学报，2017，9：1738-1745. CHEN Youdong，GUO Jiaxin，TAO Yong. Adaptive grasping strategy of robot based on Gaussian process[J]. Journal of Beijing University of Aeronautics and Astronsutics，2017，43(9)：1738-1745.
[19] 赵传松，任红格，等. 内在动机轮式倒立摆反应式认知系统[J]. 浙江大学学报(工学版)，2018，52(6)：1073-1080. ZHAO Chuansong，REN Hongge，et al. Wheeled inverted pendulum reactive cognitive system with internal motivation[J]. Journal of Zhejiang University(Engineering Science)，2018，52(6)：1073-1080.
[20] ZHANG L，ZHANG Y，LI Y. Path planning for indoor mobile robot based on deep learning[J]. Optik，2020，219：165096.
[21] 阮晓钢，张家辉，黄静，等. 一种结合内在动机理论的移动机器人环境认知模型[J]. 控制与决策，2021，3：1-7. RUAN Xiaogang，ZHANG Jiahui，HUANG Jing，et al. An environment cognition model combined with intrinsic motivation for mobile robots[J]. Control and Decision，2021，3：1-7.
[22] 苏国韶，张研，肖义龙. 高斯过程机器学习及其工程应用[M]. 北京：科学出版社，2021. SU Guoshao，ZHANG Yan，XIAO Yilong. Gaussian Process for Machine learning and Its Applications in Engineering[M]. Beijing：Science Press，2021.
[23] KOOIJMAN J D G，SCHWAB A L，MEIJAARD J P. Experimental validation of a model of an uncontrolled bicycle[J]. Multibody System Dynamics，2008，19(1/2)：115-132.