CN-121616178-B - Method and system for processing business data of narcotic drug delivery robot

CN121616178BCN 121616178 BCN121616178 BCN 121616178BCN-121616178-B

Abstract

The invention relates to the technical field of data processing, in particular to a method and a system for processing business data of an anesthetic delivery robot. The method comprises the steps of obtaining a target discrimination index value corresponding to a dead time period, and adjusting the DTW distance between an accumulated path length sequence corresponding to the dead time period in each actual delivery sub-time period and an accumulated path length sequence corresponding to a matched planned delivery path period corresponding to the actual delivery sub-time period according to the target discrimination index value to obtain a difference evaluation result between a planned delivery path and an actual delivery path of a target narcotic drug delivery task. The DTW distance between the actual conveying path and the planned conveying path is adjusted according to the target discrimination index value, so that the reliability of similarity or difference assessment of the actual conveying path and the planned conveying path of the narcotic drug conveying robot can be improved.

Inventors

BAI XUE
ZHANG JICHENG
FANG WEI
LIU BEIBEI
Qin Xiantan
ZHANG XIN

Assignees

山东第一医科大学附属省立医院(山东省立医院)
上海轻迅信息科技股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (5)

1. A method for processing business data of an anesthetic drug delivery robot, the method comprising the steps of: Acquiring a matched planning transportation path section of each actual transportation sub-time section in an actual transportation time section corresponding to a target narcotic drug transportation task, an actual position point corresponding to each actual transportation time in each actual transportation sub-time section, an accumulated path length of each actual running time, each planning position point on the matched planning transportation path section, the accumulated path length of each planning position point and a fixed stay planning position point corresponding to the target narcotic drug transportation task; acquiring a dead time period in each actual transport sub-period according to the accumulated path length of the actual running time, and obtaining a target discrimination index value corresponding to the dead time period according to a robot log under each actual running time in the dead time period, a distance between an actual position point corresponding to the actual running time in the dead time period and the fixed stay planning position point, an accumulated path length difference between adjacent actual running times in the dead time period and an angle of a vector formed by the actual position points corresponding to the adjacent actual running times; According to the target discrimination index value, adjusting the DTW distance between the accumulated path length sequence corresponding to the dead time period in each actual delivery sub-time period and the accumulated path length sequence corresponding to the matched planned delivery path period of the corresponding actual delivery sub-time period to obtain a difference evaluation result between the planned delivery path and the actual delivery path of the target narcotic drug delivery task; The method for acquiring the accumulated path length comprises the steps of marking the path length from a starting point on a planned conveying path of the target anesthetic conveying task to each planned position point on the planned conveying path as the accumulated path length corresponding to the planned position point, acquiring the planned position point closest to the actual position point corresponding to each actual running moment on the planned conveying path, marking the planned position point closest to the actual position point corresponding to the actual running moment as the latest planned position point corresponding to the actual running moment, and taking the accumulated path length of the latest planned position point corresponding to each actual running moment as the accumulated path length of the actual running moment; The method for acquiring the dead time period in each actual transport sub-time period comprises the steps of acquiring characteristic points corresponding to each actual running time in each actual transport sub-time period for any actual transport sub-time period, marking the abscissa of the characteristic points corresponding to each actual running time in the slope characteristic value sequence as a corresponding slope characteristic value after subtracting 0, and marking the result of negative correlation mapping as a corresponding slope characteristic value according to the absolute value of the slope between the characteristic points corresponding to adjacent actual running times in the actual transport sub-time period, so as to acquire a slope characteristic value sequence corresponding to the actual transport sub-time period, wherein the v-th slope characteristic value in the slope characteristic value sequence is the absolute value of the slope between the v-th actual running time and the v+1th actual running time in the actual transport sub-time period; The method for acquiring the target discrimination index value corresponding to the dead time period comprises the following steps: Obtaining a first judging index value of the dead time period according to the normal business event, a robot log under each actual running time in the dead time period and the distance between an actual position point corresponding to the actual running time in the dead time period and the fixed stop planning position point, obtaining a vector angle sequence corresponding to the dead time period, wherein the f-th vector angle in the vector angle sequence corresponding to the dead time period is the angle of a vector formed by the actual position point corresponding to the f-th actual running time in the dead time period and the actual position point corresponding to the f+1-th actual running time in the dead time period, and obtaining a second judging index value of the dead time period according to the accumulated path length difference of adjacent actual running times in the dead time period and the vector angle sequence; The method for acquiring the difference evaluation result between the planned delivery path and the actual delivery path of the target anesthetic drug delivery task comprises the following steps: Recording all dead time periods with the target discrimination index value larger than a preset discrimination threshold value as time periods to be adjusted; For any actual transporting sub-period, marking the period except the period to be adjusted in the actual transporting sub-period as a non-adjusting period, marking a sequence formed by the accumulated path lengths of all planned position points on a matched planned transporting path period of the actual transporting sub-period as a matching sequence corresponding to the actual transporting sub-period, marking the result of multiplying the DTW distance between the actual sequence corresponding to the period to be adjusted in the actual transporting sub-period and the matching sequence corresponding to the actual transporting sub-period by the adjustment degree value of the corresponding period to be adjusted, marking the actual sequence corresponding to any period as a target distance corresponding to the period to be adjusted, marking the adjustment degree value of any period to be a sequence formed by the accumulated path lengths of all actual transporting moments in the corresponding period, marking the non-adjusting period as a negative correlation mapping result of target discrimination index values corresponding to the period to be adjusted, marking the DTW distance between the actual sequence corresponding to the non-adjusting period in the actual transporting sub-period and the matching sequence corresponding to the actual sub-period as the target distance between the non-adjusting period and the target distance in the non-adjusting period; and (3) accumulating the target distances of all the actual delivery sub-time periods in the actual delivery time period, and recording the result as a difference evaluation result between the planned delivery path and the actual delivery path of the target anesthetic drug delivery task.
2. The method for processing business data of an anesthetic drug delivery robot according to claim 1, wherein the method for acquiring the matched planned delivery path segment for each actual delivery sub-period comprises: acquiring business operations specified or required in planning a target narcotic drug delivery task, and marking the business operations as necessary business operations of the target narcotic drug delivery task; Marking the position triggering the necessary business operation on the planned transport path of the target narcotic drug transport task as a necessary position point on the planned transport path, marking the time triggering the necessary business operation in the actual transport time period as a necessary time, and segmenting the actual transport time period by utilizing the necessary time in the actual transport time period to obtain each actual transport sub-time period in the actual transport time period; For any actual transporting sub-period, respectively marking two necessary time points on the actual transporting sub-period as a first time point and a second time point, marking necessary position points which are the same as the necessary business operation triggered by the first time point as necessary position points corresponding to the first time point, acquiring the necessary position point closest to the actual position point corresponding to the first time point in the necessary position points corresponding to the first time point, marking the necessary position point closest to the actual position point corresponding to the first time point as a matching position point of the first time point, wherein the acquiring method of the matching position point of the second time point is the same as the acquiring method of the matching position point of the first time point, and taking a planning transporting path section from the matching position point of the first time point to the matching position point of the second time point as a matching planning transporting path section of the actual transporting sub-period.
3. The method of claim 1, wherein the method of obtaining the first discrimination indicator value for the dead time period comprises: According to the robot logs under each actual running time in the dead time period and the normal service event, obtaining a characteristic label value corresponding to each actual running time in the dead time period, if the robot log under any actual running time contains the normal service event, taking 1 as the characteristic label value corresponding to the corresponding actual running time, otherwise, taking 0 as the characteristic label value corresponding to the corresponding actual running time; Calculating the average value of the coordinates of the actual position points corresponding to all the actual running moments in the dead time period, recording the average value as a representative position point corresponding to the dead time period, calculating the negative correlation mapping result of the distance between the representative position point of the dead time period and the fixed stop planning position point nearest to the representative position point of the dead time period, recording the negative correlation mapping result as a second characterization value corresponding to the dead time period, and recording the average value of the first characterization value of the dead time period and the second characterization value corresponding to the dead time period as the first discrimination index value of the dead time period.
4. The method of claim 1, wherein the method of obtaining the second discrimination indicator value for the dead time period comprises: And recording a set formed by absolute values of differences between all adjacent actual running time accumulated path lengths in the dead time period as an accumulated path length difference sequence corresponding to the dead time period, recording a result of negative correlation mapping after accumulating all data in the accumulated path length difference sequence as a first mapping value of the dead time period, recording a set formed by absolute values of differences between adjacent vector angles in a vector angle sequence corresponding to the dead time period as an angle difference sequence corresponding to the dead time period, recording a result of negative correlation mapping after accumulating all data in the angle difference sequence as a second mapping value of the dead time period, and recording an average value of the first mapping value and the second mapping value of the dead time period as a second discrimination index value of the dead time period.
5. A narcotic drug delivery robot service data processing system including a memory and a processor, wherein the processor executes a computer program stored in the memory to implement a narcotic drug delivery robot service data processing method as claimed in any one of claims 1 to 4.

Description

Method and system for processing business data of narcotic drug delivery robot Technical Field The invention relates to the field of data processing, in particular to a method and a system for processing business data of an anesthetic drug delivery robot. Background Currently, in order to improve safety, efficiency and supervision of transfer of narcotics in hospitals, narcotics transfer robots are introduced in the field of narcotics transfer, and are intelligent mobile devices deployed in hospitals and used for closed-loop, traceable and controlled drug transfer among operating rooms, anesthesia departments, pharmacy departments and hospitalization departments. While currently, in order to ensure delivery safety and compliance, such as to identify violations or risk behaviors such as route deviations, abnormal detours, abnormal stops, etc. in actual delivery, it is often necessary to perform a path trajectory check or to analyze and evaluate the similarity or difference between the actual delivery path and the planned delivery path of the narcotic drug delivery robot, the path trajectory check or the evaluation of the similarity or difference between the actual delivery path and the planned delivery path of the narcotic drug delivery robot belongs to one of the parts of the narcotic drug delivery robot business data processing, i.e., the narcotic drug delivery robot business data processing includes the evaluation of the similarity or difference between the actual delivery path and the planned delivery path of the narcotic drug delivery robot, the similarity or difference between the actual delivery path and the planned delivery path of the narcotic drug delivery robot is generally evaluated based on dynamic time regularity in the prior art, however, the conventional DTW aligns the sequences through elastic expansion and contraction of a time axis, and the actual running path of the narcotic drug delivery robot and the planned delivery path have obvious unique corresponding relation, in this case, the free alignment of the DTW may introduce an invalid path, so that the evaluation result has a problem of lower reliability, for example, the robot may have normal stagnation behavior, but when the robot is normally stagnated, the DTW may generate a plurality of repeated points at the same position, at this time, the points in the normal stagnation stage may be repeatedly matched to cause the accumulated distance to be suddenly increased, so that the obtained similarity or difference evaluation result may have deviation or deviation, and then the subsequent process based on the obtained similarity or difference evaluation result may have a situation of recognition error when performing the recognition of violations or risk behaviors such as route deviation, abnormal detour, abnormal stay, etc., if the normal behavior is identified as the abnormal behavior or the low-risk abnormal behavior is identified as the high-risk abnormal behavior, and therefore, when the similarity or the difference between the actual conveying path and the planned conveying path of the narcotic drug conveying robot is evaluated by using the traditional dynamic time warping, the obtained similarity or difference evaluation result has a problem of low reliability, and how to improve the reliability of the evaluation becomes a problem of a highly needed result. Disclosure of Invention In order to solve the problems, the invention provides a method and a system for processing service data of an anesthetic drug delivery robot, wherein the adopted technical scheme is as follows: in a first aspect, an embodiment of the present invention provides a method for processing service data of an anesthetic drug delivery robot, including the following steps: Acquiring a matched planning transportation path section of each actual transportation sub-time section in an actual transportation time section corresponding to a target narcotic drug transportation task, an actual position point corresponding to each actual transportation time in each actual transportation sub-time section, an accumulated path length of each actual running time, each planning position point on the matched planning transportation path section, the accumulated path length of each planning position point and a fixed stay planning position point corresponding to the target narcotic drug transportation task; acquiring a dead time period in each actual transport sub-period according to the accumulated path length of the actual running time, and obtaining a target discrimination index value corresponding to the dead time period according to a robot log under each actual running time in the dead time period, a distance between an actual position point corresponding to the actual running time in the dead time period and the fixed stay planning position point, an accumulated path length difference between adjacent actual running times in the dead time period and an angle of a vector forme