Please wait a minute...
Front. Inform. Technol. Electron. Eng.  2016, Vol. 17 Issue (6): 553-565    DOI: 10.1631/FITEE.1500269
    
Intelligent computing budget allocation for on-road trajectory planning based on candidate curves
Xiao-xin Fu, Yong-heng Jiang, De-xian Huang, Jing-chun Wang, Kai-sheng Huang
Department of Automation, Tsinghua University, Beijing 100084, China; Department of Automotive Engineering, Tsinghua University, Beijing 100084, China
Download:   PDF(0KB)
Export: BibTeX | EndNote (RIS)      

Abstract  In this paper, on-road trajectory planning is solved by introducing intelligent computing budget allocation (ICBA) into a candidate-curve-based planning algorithm, namely, ordinal-optimization-based differential evolution (OODE). The proposed algorithm is named IOODE with ‘I’ representing ICBA. OODE plans the trajectory in two parts: trajectory curve and acceleration profile. The best trajectory curve is picked from a set of candidate curves, where each curve is evaluated by solving a subproblem with the differential evolution (DE) algorithm. The more iterations DE performs, the more accurate the evaluation will become. Thus, we intelligently allocate the iterations to individual curves so as to reduce the total number of iterations performed. Meanwhile, the selected best curve is ensured to be one of the truly top curves with a high enough probability. Simulation results show that IOODE is 20% faster than OODE while maintaining the same performance in terms of solution quality. The computing budget allocation framework presented in this paper can also be used to enhance the efficiency of other candidate-curve-based planning methods.

Key wordsIntelligent computing budget allocation      Trajectory planning      On-road planning      Intelligent vehicles      Ordinal optimization     
Received: 18 August 2015      Published: 06 June 2016
CLC:  TP242.6  
Cite this article:

Xiao-xin Fu, Yong-heng Jiang, De-xian Huang, Jing-chun Wang, Kai-sheng Huang. Intelligent computing budget allocation for on-road trajectory planning based on candidate curves. Front. Inform. Technol. Electron. Eng., 2016, 17(6): 553-565.

URL:

http://www.zjujournals.com/xueshu/fitee/10.1631/FITEE.1500269     OR     http://www.zjujournals.com/xueshu/fitee/Y2016/V17/I6/553


基于候选曲线的公路轨迹规划中的智能计算量分配

目的:公路轨迹规划是辅助驾驶和无人驾驶领域中的关键问题。为解决该问题,作者针对传统方法(如势场法、RRT)在求解质量和求解效率方面的不足,提出基于候选曲线的规划算法OODE。OODE分轨迹曲线和加速度变化两部分规划轨迹,采用差分进化(DE)算法通过求解子问题计算各候选曲线的评价,然后通过比较曲线评价从候选者中选取最优曲线。DE的迭代次数越多,曲线评价越准确。本文考虑对不同曲线智能分配迭代计算量,以减少总计算量消耗,同时保证所选最优曲线以足够高的概率是真实最优曲线,从而提高OODE算法的效率。
创新点:提出基于智能计算量分配(ICBA)的轨迹规划算法框架;设计曲线评价预测模型和优质曲线选拔模型,提出基于ICBA的轨迹规划算法IOODE。
方法:基于对优质曲线迭代分配计算量的思想,设计智能计算量分配(ICBA)机制,提出基于ICBA的轨迹规划算法框架(图4);设计曲线评价预测模型(EPM)和优质曲线选拔模型(CSM),提出基于ICBA的轨迹规划算法IOODE;通过仿真分析IOODE算法的轨迹规划结果(图9、10),验证所提出计算量分配机制的有效性(图12、13)和ICBA对算法效率的提升作用(图14、表5)。
结论:本文中提出的IOODE算法与OODE算法相比,求解质量没有明显区别,但求解速度提升约20%(表5)。

关键词: 智能计算量分配,  轨迹规划,  公路规划,  智能汽车,  序优化 
[1] Wei Yang, Can-jun Yang, Ting Xu. Human hip joint center analysis for biomechanical design of a hip joint exoskeleton[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(8): 792-802.
[2] Qiang Liu, Jia-chen Ma. Subspace-based identification of discrete time-delay system[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(6): 566-575.
[3] Feng-yu Zhou, Xian-feng Yuan, Yang Yang, Zhi-fei Jiang, Chen-lei Zhou. A high precision visual localization sensor and its working methodology for an indoor mobile robot[J]. Front. Inform. Technol. Electron. Eng., 2016, 17(4): 365-374.
[4] Qian-shan Li, Rong Xiong, Shoudong Huang, Yi-ming Huang. Building a dense surface map incrementally from semi-dense point cloud and RGB images[J]. Front. Inform. Technol. Electron. Eng., 2015, 16(7): 594-606.
[5] Xin Ma, Ya Xu, Guo-qiang Sun, Li-xia Deng, Yi-bin Li. State-chain sequential feedback reinforcement learning for path planning of autonomous mobile robots[J]. Front. Inform. Technol. Electron. Eng., 2013, 14(3): 167-178.
[6] Hao-jie Zhang, Jian-wei Gong, Yan Jiang, Guang-ming Xiong, Hui-yan Chen. An iterative linear quadratic regulator based trajectory tracking controller for wheeled mobile robot[J]. Front. Inform. Technol. Electron. Eng., 2012, 13(8): 593-600.
[7] Yuan-hui Zhang, Wei Wei, Dan Yu, Cong-wei Zhong. A tracking and predicting scheme for ping pong robot[J]. Front. Inform. Technol. Electron. Eng., 2011, 12(2): 110-115.