家禽智慧养殖农场自主投料机器人任务路径规划算法  

An Algorithm for Autonomous Feeding Robot Task Path Planning in Poultry Smart Farm

作　　者：张燕军[1] 孙卫明 郭阳阳 杨坚 吴伟伟[1] 缪宏[1] 张善文[1] ZHANG Yan-jun;SUN Wei-ming;GUO Yang-yang;YANG Jian;WU Wei-wei;MIAO Hong;ZHANG Shan-wen(College of Mechanical Engineering,Yangzhou University,Yangzhou 225127,China)

机构地区：[1]扬州大学机械工程学院,扬州225127

出　　处：《科学技术与工程》2023年第8期3207-3215,共9页Science Technology and Engineering

基　　金：国家自然科学基金(32072785);江苏省农业自主创新基金(CX(19)3010,CX(22)3032)。

摘　　要：为了解决家禽智能养殖场饲养环境差、喂养不及时、劳动力需求大等问题,投料机器人的开发迫在眉睫,而路径规划算法的研究是开发饲养机器人的重要环节。饲喂机器人的能耗是路径规划过程中关注的重要内容之一。投料机器人的总重量在投料过程中一直在变化,以最短路径行驶并不意味着其能耗最小,因此需要找到一条最合适的路径而不是最短的路径来使得投料机器人行驶能耗最小。针对以上问题,提出对一定数量的缺料饲料桶采用分支定界算法规划出一条能耗最小的路径;通过最短边集优先选取结合序列不等式方法计算出投料机器人行驶的能耗下界,通过基于Christofides启发式算法计算出投料机器人行驶的能耗上界,同时计算并证明能耗上界与最优能耗的最坏比界。通过在投料机器人的工控机上的运算结果表明:相较于其他求解能耗上下界的方法,所提出方法的求解结果更加精确,能够保证分支定界算法的求解速度更快,同时分支定界算法能快速地求出投料机器人的能耗最小的行驶路径,相较于其他算法而言求解能力更强。In order to solve the problems of poor feeding environment,untimely feeding and high labor demand in poultry smart farms,the development of feeding robots is imminent,while the research on path planning algorithms is an important part of developing feeding robots.The energy consumption of the feeding robot is one of the important elements of concern in the process of path planning.The total weight of the feeding robot changes all the time during the feeding process,and traveling with the shortest path does not mean its energy consumption is the smallest,therefore,it is necessary to to find a most suitable path instead of the shortest path to make the feeding robot travel with the smallest energy consumption.For the above problems,a branch-and-bound algorithm was proposed to obtain a path with minimum energy consumption.The lower bound of energy consumption for the feeding robot was obtained by the approach of preferred selection of the set of shortest edges combined with the sequence inequality,the upper bound of energy consumption for the feeding robot was obtained by the Christofides s heuristic algorithm,and the worst ratio between the upper bound and the optimal energy consumption was also calculated and proved.The results of the calculations on the industrial control computer of the feeding robot show that compared with other methods for calculating the upper and lower bounds of energy consumption,the solution results of the proposed method are more accurate,and it could ensure the calculation speed of branch and bound algorithm.The proposed branch and bound algorithm can calculate the minimum energy consumption path speedily,which is better than other algorithms.

关 键 词：智慧农场 投料机器人 路径规划 分支定界 能耗优化 精准农业 

分 类 号：S115[农业科学—农业基础科学]