CN111836199B - Indoor navigation dynamic obstacle avoidance and path finding method - Google Patents

Indoor navigation dynamic obstacle avoidance and path finding method Download PDF

Info

Publication number
CN111836199B
CN111836199B CN202010486710.5A CN202010486710A CN111836199B CN 111836199 B CN111836199 B CN 111836199B CN 202010486710 A CN202010486710 A CN 202010486710A CN 111836199 B CN111836199 B CN 111836199B
Authority
CN
China
Prior art keywords
layer
room
path
grid
network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010486710.5A
Other languages
Chinese (zh)
Other versions
CN111836199A (en
Inventor
刘骝
李博峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tongji University
Original Assignee
Tongji University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tongji University filed Critical Tongji University
Priority to CN202010486710.5A priority Critical patent/CN111836199B/en
Publication of CN111836199A publication Critical patent/CN111836199A/en
Application granted granted Critical
Publication of CN111836199B publication Critical patent/CN111836199B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/024—Guidance services
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations
    • G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/005—Discovery of network devices, e.g. terminals

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Databases & Information Systems (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Navigation (AREA)

Abstract

The invention relates to an indoor navigation dynamic obstacle avoidance and path finding method, which comprises the following steps: (a) modeling an indoor passable area based on a double-layer mixed data model according to an indoor map in a target building, wherein the model is divided into a first-layer topological network and a second-layer grid network; (b) dividing a first-layer topological network of a building; (c) dividing a second layer of grid network of the building; (d) dividing the barrier; (e) counting the congestion degree of each room by adopting the method in the step (d), and storing the congestion degree as attribute data on the nodes of the first-layer topological network in the step (b); (f) calculating an optimal obstacle avoidance path for the specified user according to the input current user position and the target position; (g) determining whether to adjust and calculate a new optimal obstacle avoidance path according to the crowd density distribution monitored in real time; the invention obtains the traversable path in real time, and greatly improves the accuracy of navigation path calculation.

Description

Indoor navigation dynamic obstacle avoidance and path finding method
Technical Field
The invention relates to the field of indoor navigation of pedestrians, in particular to a dynamic obstacle avoidance and path finding method for indoor navigation.
Technical Field
Indoor navigation services are now being developed in great lengths, and such technologies lay the foundation for indoor location-based services. As an important component of providing navigation information, path planning (i.e., path finding) is one of the core tasks of indoor navigation. For pedestrian navigation in a conventional scene, an indoor navigation network is used for abstractly expressing the space where a user is located and the mutual position relation of the space, and is used for indoor routing calculation. The mainstream calculation method is to generate an indoor traffic network and calculate a path in advance according to an indoor map, or to label the path in advance on an existing indoor road network.
Aiming at indoor navigation of pedestrians, different solutions for indoor navigation routing at home and abroad are provided at present. Map manufacturers such as google, hundredth, and high grade have been able to provide vector format indoor maps for some public buildings, but they often employ fixed indoor navigation networks or do not provide indoor networks. These corresponding paths are all indicated paths oriented to static environments, and the calculation of the optimal path is often based on the shortest distance. Although a few technical solutions for dynamic update of an indoor navigation network exist at present, the description of the conditions and specific navigation effects for updating the road network is not clear enough, and the applied scene is small (such as an office building).
In general, most of the current indoor pedestrian navigation networks are mainly constructed for static indoor environments, static obstacles are often ignored, and therefore the corresponding routing calculation result provides an indication path without considering obstacle avoidance. In addition, the existing navigation network construction and routing methods are lack of description and expression of dynamic obstacles. The dynamic obstacle refers to a movable obstacle or an obstacle with a boundary changing along with time (such as the movement of a crowd and an equipment group), and accurate routing calculation needs to adjust a planned path in real time according to the dynamic obstacle, for example, the congestion degree of indoor crowds in different time periods can influence the navigation path selection of a user. Because the influence of dynamic obstacles is generally ignored, the existing indoor navigation model and routing method cannot completely reflect the dynamic change in the building, and therefore the accuracy and the usability of the indoor navigation model and the routing method are reduced.
Disclosure of Invention
Aiming at the problems, the invention aims to provide an indoor navigation dynamic obstacle avoidance and path finding method based on a double-layer mixed data model, and solves the problems that dynamic information is difficult to update accurately in real time in the construction of the existing indoor navigation model, dynamic obstacles are less considered in global path planning and the like.
In order to achieve the purpose, the invention adopts the following technical scheme:
an indoor navigation dynamic obstacle avoidance and path finding method based on a double-layer hybrid model is characterized by comprising the following steps:
(a) and modeling an indoor passable area based on the double-layer mixed data model according to the indoor map in the target building. The model is divided into a first layer of topological network and a second layer of grid network, wherein the first layer of topological network expresses the connectivity of a room; and the second layer of grid network consists of regular grid units and expresses the space and the state inside the room.
(b) And dividing a first-layer topological network of the building, and dynamically maintaining a connectivity graph (a topological structure for expressing the accessibility of the rooms) of each room. Wherein each node represents a room and wherein each node represents a room,each edge represents the connectivity of two rooms. The first layer topology network is represented as: gt＝{Nt,Et},Nt＝{n1,n2…nkK is the room number }, Et＝{(ni,nj)|(ni,nj∈Nt)∩(niAnd njConnected), 1 is less than or equal to i, j is less than or equal to k, and i is not equal to j. To EtEach side (n) ofi,nj) Store the distance d thereofijAnd congestion degree bij。
(c) And dividing a second layer grid network of the building, and adopting a regular grid model for each room to accurately reflect the influence of static obstacles and dynamic obstacles. The size of the grid unit is determined according to actual requirements (such as 1-3 m). The second layer of the grid network is represented as:
Figure BDA0002519498950000021
room
Figure BDA0002519498950000022
Vd＝{v1,v2,…vpP is nkNumber of inner grids }, Ed＝{(vi,vj)|(vi,vj∈Vd)∩(viAt vjIn the 8-neighborhood), 1 is less than or equal to i, j is less than or equal to p, and i is not equal to j. For VdEach grid v inpStoring the information O it occupiespAnd density of persons therein Dp. If v ispIs completely occupied, then Op0, otherwise 1; dpIn a person/m2And (4) showing.
(d) Dividing barriers, including static and dynamic barriers, to which the indoor pedestrian flow belongs in the present application;
collecting mobile equipment signals indoors by using Wi-Fi scanning equipment, analyzing and updating the crowd density and distribution in the corresponding room in real time, and outputting the real-time barrier distribution result in the second layer grid network in the step (c);
according to the crowd density in the grids in the rooms, the congestion degree of each room is divided into: smooth, light congestion, and severe congestion.
(e) Counting the congestion degree of each room by adopting the method in the step (d), and storing the congestion degree as attribute data on the nodes of the first-layer topological network in the step (b);
the distance between the passage openings (e.g., door, stairway opening, elevator opening, etc.) of each room is calculated in advance and stored as attribute data on the connected side of the first-level topology network in step (b).
(f) And calculating the optimal obstacle avoidance path (with the shortest length or time) for the specified user according to the input current user position and the target position. The dynamic obstacle avoidance and path finding calculation depends on the double-layer mixed data model, and the calculation efficiency is optimized in a layered calculation mode. The model is expressed by a set theory notation, wherein a graph model G of a first-layer topological network is { N, E }, and a building is divided into a room set N according to boundaries such as walls and the like, and the room set N is { N }1…Nt}, node NtCorresponding to the t-th room. Establishing a set E of all connected edges according to connected roomsi,jE belongs to E; node N for each roomk(k<T) creating a subgraph (i.e. a second-level grid model) Gk＝{Nk,EkTherein of
Figure BDA0002519498950000033
Comprising a room NkAll the unit nodes (p in total) of the middle grid model, and the connection paths among the unit nodes are
Figure BDA0002519498950000034
First floor node N through k roomskAnd corresponding second layer structure GkIn the updating of GkSimultaneous updating of first-tier topology nodes N for real-time informationkAnd realizing linkage between double-layer data.
A global room sequence between a starting point and an end point is first calculated for a user on a first-tier topology network G. Firstly, calculating the intermediary centrality C (v) of all nodes in G according to an intermediary centrality formula:
Figure BDA0002519498950000031
n in the formula (1)st(v) Denotes the number of shortest paths through node v between start point s and end point t, and nstIndicating the number of shortest paths from s to t. And highlighting the passing junction nodes on the first-layer topological network by adopting the parameter C (v). Passing the intermediary centrality of each node in G to the corresponding edge EijAnd according to EijDistance d ofijAnd current congestion degree bijCalculating the traffic weight P of each edge by using a weighting methodij:
Figure BDA0002519498950000032
W in the formula (2)1Preference weight, w, for Medium centrality C2Preference weight for distance, w3Preference for congestion tolerance. Aiming at the obstacle avoidance requirement, setting w3>w1>w2Namely, the path calculation firstly re-avoids the obstacle, secondly is the junction area, and finally is the distance. According to the current traffic weight P of all edges of G, calculating the path (min sigma P) of the minimum P by using a classic shortest path algorithm Dijkstra methodij) A first layer path (i.e., a room sequence) is obtained over the topological network G.
Thereafter, in each room included in the sequence, a local dynamic obstacle avoidance path is accurately calculated based on the second-layer grid network. The boundary of the static obstacle is discretized, and the corresponding grid cell is set as impassable. Meanwhile, according to the real-time data, the current crowd density of other grid units is calculated, and the maximum influence range of the dynamic obstacle is determined. Defining a degree of obstruction B for a grid cellk,tComprises the following steps:
Bk,t＝Nk,t/Uk (3)
Nk,tthe number of people detected by the k unit at the time t; u shapekAn upper limit on the number of people that the unit can accommodate. A blocking degree threshold BH is set, and a mesh cell exceeding the threshold is set as impassable. Because the uncertainty of the dynamic position affects the blocking distance in different gridsAccuracy of degree, and therefore such uncertainty is expressed by calculating the position transition probability:
Figure BDA0002519498950000041
in the formula (4), G represents a grid cell participating in calculation. Kappa is the normalized coefficient, xk,tGrid cell k, z representing time ttWhich represents the position estimate at the time t,
Figure BDA0002519498950000042
is an estimate. By counting P in each grid cellk,tB of each unit is obtainedk,tA probability value.
B for obtaining all grid units on the second floor of the current roomk,tAfter the value is obtained, the path in the current room is searched by the a-x algorithm (the existing algorithm, which is the prior art). Adding B to the heuristic function H (m) of a certain grid cell m at time t according to the cost function F (m) ═ G (m) + H (m) in the A-algorithmk,tBased on the constraint condition min (g (m) + h (m)), a passage path for the obstacle change in the current room grid is calculated. According to the process, after grid models in all rooms in the first-layer path are calculated, the second-layer paths are connected in sequence to obtain a complete dynamic obstacle avoidance path.
(g) And updating the congestion degree on the selected path according to the crowd density distribution monitored in real time, and determining whether to adjust and calculate a new optimal obstacle avoidance path according to the current position of the user. And if the congestion degree in the current room is increased according to the monitoring result, updating the dynamic information of the double-layer model and calculating the current optimal obstacle avoidance path.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention provides an indoor navigation dynamic obstacle avoidance and path finding method based on a double-layer hybrid data model. The traditional indoor path planning or path finding method does not fully consider the influence of dynamic environment change on the path finding result. Compared with the traditional pedestrian navigation routing method, the method can specially process the influence of dynamic obstacles (such as pedestrian flow) in indoor navigation on indoor passable space, and dynamically updates the navigation network in real time, thereby assisting in realizing accurate dynamic obstacle avoidance calculation, obtaining passable paths in real time and greatly improving the accuracy of navigation path calculation.
(2) In order to enable the navigation network to flexibly adapt to indoor environment changes, the navigation path planning method with good feasibility is provided based on a double-layer mixed data model. The traditional multi-level indoor model structure (such as 'floor-area-room') is built with more levels, such as G ═ N, E },
Figure BDA0002519498950000051
wherein m is the mth layer. Aiming at a navigation routing scene, although the structure can theoretically express an indoor space, the expression is easy to be complicated, and a corresponding specific multilayer linkage routing calculation method is lacked. Compared with the method, the method has the corresponding specific navigation routing calculation method, reduces the cognitive redundancy of the pedestrian user to the building (only the room needing to pass through is concerned), and can ensure that the navigation network is updated in time and the route is adjusted in real time.
(3) Different from the traditional indoor path finding method, the invention provides an indoor dynamic obstacle avoidance path planning method based on a double-layer mixed data model, which comprises the following steps: calculating a room sequence based on the first layer of topological network, and avoiding simultaneous calculation of large-scale grid units; and performing accurate path searching calculation based on the local grid network of the second layer in the selected room. Therefore, the high efficiency of calculation is effectively ensured, and the real-time performance of the influence of dynamic environment change on the path searching is fully considered. The invention considers uncertain dynamic barriers in pedestrian indoor navigation and expands the application of indoor navigation routing into real-time calculation.
Drawings
FIG. 1 is a schematic diagram of an embodiment of the present invention.
FIG. 2 is a schematic diagram of a topological network of the first floor room interconnectivity in the two-tier hybrid data structure of the present invention.
Fig. 3 is a schematic diagram illustrating an abstract path computation on a first-layer navigation network according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of the passable areas in each room on the second tier network in accordance with an embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in detail with reference to the accompanying drawings so that the advantages and features of the invention can be more easily understood by those skilled in the art, and the scope of the invention will be clearly and clearly defined.
The present invention will be described in further detail with reference to examples, but the embodiments of the present invention are not limited thereto.
As shown in fig. 1, an indoor navigation dynamic obstacle avoidance and path finding method based on a double-layer hybrid data model includes the following steps:
the method comprises the following steps of 1, modeling an indoor environment by adopting a double-layer mixed data model based on an indoor map of a building, expressing the connectivity of an indoor space by using a topological network model, and increasing the detailed semantics and state in the expression space by using a grid model as an embodiment but not a limitation.
And 2, in the first layer of the double-layer hybrid data model, constructing a connectivity network (figure 2) among rooms, and dynamically maintaining and updating the connectivity of each room. Its nodes represent each room (office, corridor, stairwell, etc.) within the building and the edges represent the accessibility of the rooms. The edge of the network stores attribute values such as distance and congestion degree to express the current state.
And step 3: in the second layer, a grid model is built within each room, expressing the current state of the navigable areas and other locations within each room (fig. 4). Each grid cell stores information on whether it is currently occupied, including the density of people in the grid, etc.
And 4, step 4: and determining the number and the positions of the mobile devices according to the scanning result of the Wi-Fi device at the current moment, and inputting the personnel density to the corresponding grid. The higher the density of the people in the grid is, the higher the congestion degree is. A threshold level of obstruction is set and grid cells exceeding the threshold are considered to be impassable.
And 5: and updating the first-layer network according to the current dynamic barrier information. And counting the number t of grid units exceeding a threshold value in each room, and calculating the ratio R to the total number n of grid units in the room, wherein the ratio is t/n. And if the proportion R exceeds a certain threshold value, updating the congestion degree of the corresponding edge on the first-layer network (for example, changing into 'severe congestion').
Step 6: and performing hierarchical calculation based on a double-layer mixed data model according to the input current position and the target position of the user. The invention abstracts the building space hierarchy into a compact 'building-room' structure, and the floor information can be used as the property of the room. Therefore, multiple division of indoor scenes is avoided, and the cognitive burden of pedestrian users is avoided.
And 7: and (4) calculating an abstract path on the first-layer topological network, namely performing global search on the whole building and determining a room sequence to be passed through. The path includes the indoor room and stairs/escalators/elevators etc. that the current user needs to pass through (fig. 3). The routing strategy adopted is as follows: first selecting a path in an unobstructed area; avoiding heavily congested areas. The path is searched from low to high with the R value of each room. E.g., R > 50%, the room is not considered when the first layer seeks. From a starting point, a sequence of rooms that can be traveled for a user is calculated on a first tier network (fig. 3).
And 8: and calculating an accurate dynamic obstacle avoidance geometric path on the second layer of grid network. The second layer is a local search, searching for the exact geometric path on all grid cells in the specified room (fig. 4). If only static obstacles exist in the room, directly calculating an obstacle avoidance path; if dynamic obstacles exist in the room, determining the maximum influence range of the grids according to the current crowd density of the grid units (figure 4), and carrying out obstacle avoidance calculation according to the influence range.
And combining the room sequence (abstract path) obtained by the first layer of calculation, completing the second layer of path finding calculation in each room in sequence, and splicing all paths to obtain an accurate dynamic obstacle avoidance geometric path (fig. 4).
And step 9: and according to the currently calculated accurate obstacle avoidance path, determining whether the vehicle can pass smoothly according to the plan in each room. And if the path passes through, moving to the target position according to the planned path.
Step 10: and if the user moves forward in the current room according to the planned path and the encountered dynamic obstacle obviously hinders the user from moving, updating the personnel density on the second-layer raster model and updating the congestion information of the first-layer topological network according to the real-time data of the Wi-Fi scanning equipment at the moment. And replanning a new optimal obstacle avoidance path based on the double-layer mixed data model by taking the current position as a starting point (namely, repeating the steps 4-8).

Claims (1)

1. An indoor navigation dynamic obstacle avoidance and path finding method based on a double-layer hybrid model is characterized by comprising the following steps:
(a) modeling an indoor passable area in a target building according to an indoor map and based on a double-layer mixed data model, wherein the model is divided into a first-layer topological network and a second-layer grid network, and the first-layer topological network expresses the connectivity of rooms; the second layer of grid network consists of regular grid units and expresses the space and the state inside the room;
(b) dividing a first-layer topological network of a building, and dynamically maintaining a connectivity graph of each room, wherein each node represents a room, and each edge represents the connectivity of two rooms; the first layer topology network is represented as: gt＝{Nt,Et},Nt＝{n1,n2…nkK is the room number }, Et＝{(ni,nj)|(ni,nj∈Nt)∩(niAnd njConnected), 1 is less than or equal to i, j is less than or equal to k, and i is not equal to j, for EtEach side (n) ofi,nj) Store the distance d thereofijAnd congestion degree bij；
(c) Dividing a second layer of grid network of the building, adopting a regular grid model for each room to accurately reflect the influence of static obstacles and dynamic obstacles, and determining the size of grid units according to actual requirements; the second layer of the grid network is represented as:
Figure FDA0003017054400000011
room nk,
Figure FDA0003017054400000012
Vd＝{v1,v2,…vpP is nkNumber of inner grids }, Ed＝{(vi,vj)|(vi,vj∈Vd)∩(viAt vjIn 8-neighborhood) 1 ≦ i, j ≦ p and i ≠ j, for VdEach grid v inpStoring the information O it occupiespAnd density of persons therein DpIf v ispIs completely occupied, then Op0, otherwise 1; dpIn a person/m2Represents;
(d) dividing barriers including static and dynamic barriers, wherein indoor pedestrian flow belongs to the dynamic barriers;
acquiring mobile equipment signals indoors by using Wi-Fi scanning equipment, analyzing and updating the personnel density and distribution in the corresponding room in real time, and outputting the real-time barrier distribution result in the second layer grid network in the step (c);
according to the personnel density in the grid in the rooms, the congestion degree of each room is divided into: smooth, light congestion and severe congestion;
(e) counting the congestion degree of each room by adopting the method in the step (d), and storing the congestion degree as attribute data on the nodes of the first-layer topological network in the step (b);
pre-calculating the distance between the channel ports of each room, and storing the distance as attribute data on the connection edge of the first-layer topology network in the step (b);
(f) calculating an optimal obstacle avoidance path for the specified user according to the input current user position and the target position; the dynamic obstacle avoidance and path finding calculation depends on the double-layer mixed data model, the calculation efficiency is optimized in a layered calculation mode, the model is expressed by a set theory symbol, a graph model G of a first-layer topological network is { N, E }, and a building is divided into a room set N according to the boundaries such as walls and the like1…Nt}, node NtCorresponding to the t-th room; according to connected roomsSet E of all connected edgesi，jE belongs to E; node N for each roomkBuild a subgraph, k<＝t，Gk＝{Nk，EkTherein of
Figure FDA0003017054400000021
Comprising a room NkThe number of all unit nodes of the middle grid model is p, and the connection path between the unit nodes is
Figure FDA0003017054400000022
First level node Nk and corresponding second level structure G through k roomskIn the updating of GkSimultaneous updating of first-tier topology nodes N for real-time informationkThe state of (2), realize the linkage between the double-deck data;
firstly, calculating a global room sequence between a starting point and an end point on a first-layer topological network G for a user; firstly, calculating the intermediary centrality C (v) of all nodes in G according to an intermediary centrality formula:
Figure FDA0003017054400000023
n in the formula (1)st(v) Denotes the number of shortest paths through node v between start point s and end point t, and nstRepresenting the number of shortest paths from s to t; highlighting a passing junction node on the first-layer topological network by adopting a parameter C (v); passing the intermediary centrality of each node in G to the corresponding edge EijAnd according to EijDistance d ofijAnd current congestion degree bijCalculating the traffic weight P of each edge by using a weighting methodij：
Figure FDA0003017054400000024
W in the formula (2)1Preference weight, w, for Medium centrality C2Preference weight for distance, w3Preference for tolerance to congestion level; aiming at the obstacle avoidance requirement, setting w3＞w1＞w2Namely, the path calculation firstly re-avoids the obstacle, secondly is a junction area, and finally is the distance; according to the current traffic weight P of all edges of G, calculating the path (min sigma P) of the minimum P by using a classic shortest path algorithm Dijkstra methodij) Obtaining a first layer path on the topology network G;
then, in each room contained in the sequence, accurately calculating a local dynamic obstacle avoidance path based on the second layer grid network; discretizing the boundary of the static barrier, and setting the corresponding grid cell as impassable; meanwhile, according to the real-time data, calculating the current personnel density of other grid units, and determining the maximum influence range of the dynamic barrier; defining a degree of obstruction B for a grid cellk，tComprises the following steps:
Bk，t＝Nk，t/Uk (3)
Nk，tthe number of people detected by the k unit at the time t; u shapekAn upper limit on the number of people that the unit can accommodate; setting a blockage degree threshold BH, and setting the grid cells exceeding the threshold as impassable; since uncertainty in dynamic position affects the accuracy of the degree of obstruction in different grids, such uncertainty is expressed by calculating position transition probabilities:
Figure FDA0003017054400000031
in the formula (4), G represents a grid cell participating in calculation; kappa is the normalized coefficient, xk，tGrid cell k, z representing time ttWhich represents the position estimate at the time t,
Figure FDA0003017054400000032
is an estimate; by counting P in each grid cellk，tB of each unit is obtainedk，tA probability value;
b for obtaining all grid units on the second floor of the current roomk，tAfter the value is obtained, searching a path in the current room by adopting an A-star algorithm; according to the cost function F (in A) algorithmm) G (m) + H (m), and at time t, for a certain grid cell m, B is added to its heuristic function H (m)k，tCalculating a passing path which is changed for the barrier in the current room grid according to the constraint condition min (G (m) + H (m)); according to the process, after grid models in all rooms in the first-layer path are calculated, the second-layer paths are connected in sequence to obtain a complete dynamic obstacle avoidance path;
(g) updating the congestion degree on the selected path according to the personnel density distribution monitored in real time, and determining whether to adjust and calculate a new optimal obstacle avoidance path according to the current position of the user; and if the congestion degree in the current room is increased according to the monitoring result, updating the dynamic information of the double-layer model and calculating the current optimal obstacle avoidance path.
CN202010486710.5A 2020-06-01 2020-06-01 Indoor navigation dynamic obstacle avoidance and path finding method Active CN111836199B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010486710.5A CN111836199B (en) 2020-06-01 2020-06-01 Indoor navigation dynamic obstacle avoidance and path finding method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010486710.5A CN111836199B (en) 2020-06-01 2020-06-01 Indoor navigation dynamic obstacle avoidance and path finding method

Publications (2)

Publication Number Publication Date
CN111836199A CN111836199A (en) 2020-10-27
CN111836199B true CN111836199B (en) 2021-06-29

Family

ID=72897515

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010486710.5A Active CN111836199B (en) 2020-06-01 2020-06-01 Indoor navigation dynamic obstacle avoidance and path finding method

Country Status (1)

Country Link
CN (1) CN111836199B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112286211A (en) * 2020-12-28 2021-01-29 上海交大智邦科技有限公司 Environment modeling and AGV path planning method for irregular layout workshop
CN113009910A (en) * 2021-02-09 2021-06-22 上海同陆云交通科技有限公司 Sidewalk detection robot system and sidewalk detection method
CN114545922A (en) * 2021-12-28 2022-05-27 美的集团(上海)有限公司 Path planning method, electronic device and computer storage medium
CN114357106A (en) * 2022-03-21 2022-04-15 广东省科学院广州地理研究所 Indoor path planning method and device
CN115826011B (en) * 2023-01-06 2023-05-05 北京讯腾智慧科技股份有限公司 RTK (real-time kinematic) method and device for dynamic grid network of image layer
CN116600252B (en) * 2023-07-17 2023-09-15 深圳市艾米艺术设计有限公司 Room route guiding method and device based on wireless hot spot and computer equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103576686A (en) * 2013-11-21 2014-02-12 中国科学技术大学 Automatic guide and obstacle avoidance method for robot
CN103792846A (en) * 2014-02-18 2014-05-14 北京工业大学 Robot obstacle avoidance guiding method based on Skinner operating condition reflection principle
CN110703747A (en) * 2019-10-09 2020-01-17 武汉大学 Robot autonomous exploration method based on simplified generalized Voronoi diagram

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103576686A (en) * 2013-11-21 2014-02-12 中国科学技术大学 Automatic guide and obstacle avoidance method for robot
CN103792846A (en) * 2014-02-18 2014-05-14 北京工业大学 Robot obstacle avoidance guiding method based on Skinner operating condition reflection principle
CN110703747A (en) * 2019-10-09 2020-01-17 武汉大学 Robot autonomous exploration method based on simplified generalized Voronoi diagram

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
面向移动机器人的混合路径规划实现;包丽;《东南大学》;20191231;全文 *

Also Published As

Publication number Publication date
CN111836199A (en) 2020-10-27

Similar Documents

Publication Publication Date Title
CN111836199B (en) Indoor navigation dynamic obstacle avoidance and path finding method
Konolige et al. Navigation in hybrid metric-topological maps
Lin et al. Intelligent generation of indoor topology (i-GIT) for human indoor pathfinding based on IFC models and 3D GIS technology
Nasir et al. Prediction of pedestrians routes within a built environment in normal conditions
CN111123198B (en) User positioning and navigation method and system in building
Liu et al. A semantic data model for indoor navigation
Kannan et al. Predictive indoor navigation using commercial smart-phones
CN115713856A (en) Vehicle path planning method based on traffic flow prediction and actual road conditions
Vanclooster et al. Evaluating suitability of the least risk path algorithm to support cognitive wayfinding in indoor spaces: an empirical study
CN107655490B (en) Hot spot path discovery method based on mobile user track segmentation and hottest search
CN115355922A (en) Travel path planning method and system based on improved ant colony algorithm
CN113358118B (en) End-to-end autonomous navigation method for indoor mobile robot in unstructured environment
CN117371760B (en) Layered passenger ship personnel emergency evacuation method considering deadline and congestion relief
Wang et al. Combining building and behavior models for evacuation planning
Cai et al. A novel vector-based dynamic path planning method in urban road network
CN112036621B (en) Method and device for setting guiding mark in transportation junction
Zhou et al. An indoor route planning method with environment awareness
Richter Adaptable path planning in regionalized environments
CN113553357B (en) HW-Louvain-based urban public transport network divisible spatial community detection method
CN112229408B (en) Three-dimensional indoor multi-floor pedestrian route planning method
CN108731688A (en) Air navigation aid and device
Song et al. A grid-based graph data model for pedestrian route analysis in a micro-spatial environment
Kolomvatsos et al. Semantic location based services for smart spaces
CN113159433B (en) Dynamic navigation path searching method for integrated indoor mixed three-dimensional road network
CN117168466B (en) Escape route planning method, escape route planning device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant