CN101719285A - Collision prevention method for multi-layered virtual communities - Google Patents
Collision prevention method for multi-layered virtual communities Download PDFInfo
- Publication number
- CN101719285A CN101719285A CN200910265430A CN200910265430A CN101719285A CN 101719285 A CN101719285 A CN 101719285A CN 200910265430 A CN200910265430 A CN 200910265430A CN 200910265430 A CN200910265430 A CN 200910265430A CN 101719285 A CN101719285 A CN 101719285A
- Authority
- CN
- China
- Prior art keywords
- collision
- collision prevention
- group
- speed
- virtual community
- 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.)
- Granted
Links
- 230000002265 prevention Effects 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004088 simulation Methods 0.000 claims abstract description 12
- 230000003068 static effect Effects 0.000 claims abstract description 12
- 230000008859 change Effects 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 abstract description 8
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 abstract 1
- 230000000638 stimulation Effects 0.000 abstract 1
- 230000006399 behavior Effects 0.000 description 19
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000008447 perception Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 244000144992 flock Species 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005381 potential energy Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a collision prevention method for multi-layered virtual communities, which relates to the technical field of the animation of virtual communities. The method mainly provides collision prevention technology of the multi-layered communities which reduces collision frequency and improves the efficiency in the process of community simulation so as to reduce the complexity of collision prevention simulation of the communities. The method comprises the steps of: performing spatial division according to the positions of static obstacles; performing grouping, the stimulation of groups and collision prevention among individuals in the groups on the virtual communities in a walkable area; determining collision types and performing collision prevention through collision forecast among the groups; and evaluating and relocating the collision prevention, and performing local optimization according to an evaluation result to generate global collision prevention speed. The method can effectively realize the collision detection and collision avoidance of virtual crowds in the animation of the virtual communities.
Description
Technical field
The present invention relates to virtual community cartoon technique field, refer in particular to a kind of collision prevention method of multi-level virtual community.
Background technology
Along with the continuous development of computer animation, the real-time simulation of colony's animation is also being made significant headway in recent years as a branch of computer animation.Colony's cartoon technique has been widely used in disaster scene simulations such as city planning, film animation, recreation creation, training system and fire, earthquake and has assisted among the commander.
Because in the simulation and simulation process of colony, no matter in action, the quantity of colony still will carry out all that very complicated algorithm and step are controlled and simulated the aspect such as playing up, this will cause the time of algorithm and space complexity bigger, so in colony's collision prediction and collision are avoided, should find a kind of method of simple possible to reduce complexity.Wherein single step detection method is relatively use always a kind of; this method is a kind of based on the method for recycling static collision detection on given trace; promptly in the movement of objects process, track is divided into a lot of time steps; all carry out static collision detection in each step (several specific time points on the movement locus of object), judge between the moving object whether bump.Obviously the step-length of sampling has directly determined algorithm accuracy and complexity.If sampling interval is excessive, some collision may detect less than; If but sampling interval is too small, calculation cost is too big.This method is not suitable for the system of real-time simulation, mainly be step-length determine will be according to the density of colony, factors such as environment determine, when the step-length of determining when not being very accurate, algorithm complex then is bigger.Based on the detection method of grid is at first grid dividing to be become very little unit, and each is individual independent among a grid, avoids with the collision of other individualities by the position of individuality grid at place when certain frame.The social force model then is by judging that two distances between the individuality produce a kind of repulsive force and simulate individual behavior, and these several collision detection and the method for avoiding all design at the behavior of a certain colony, promptly are not method in common.For example general and potential energy field, density field etc. the combine planning of carrying out the path and collision prevention based on the detection of grid.Collision detection and avoiding method based on visual range then are the vision system of elder generation by the anthropomorphic dummy, with information such as vision radius and radians the grid individuality on every side in the vision is carried out perception, if perceive individual existence, then carrying out corresponding collision prevention measure, this method only detects the individuality in the visual range, need not detect each individuality, the number of times of detection obviously reduces, but the process of perception more complicated then.
Summary of the invention
For addressing the above problem, the collision prevention method of a kind of multi-level virtual community of the present invention aims to provide a kind of technology of the collision prevention of colony by different level that can reduce collision frequency, raise the efficiency.
The technical solution adopted in the present invention comprises following operation: step 1, come virtual community is divided into groups according to spatial division; Step 2, the simulation of group and the collision of organizing between the inner individuality are avoided; Step 3, the collision prediction between group and the group; Step 4 is determined the crash type between group and the group; Step 5 is taked the collision prevention measure according to crash type; Step 6, the assessment of collision prevention and reorientation; Step 7 is carried out local optimum according to assessment result, produces the collision prevention speed of the overall situation.
Position according to the static-obstacle thing in the described step 1 is walkable region and non-walkable region with spatial division, in the certain distance scope in same walkable region Different Individual is divided into one group.
Crash type in described step 4 and the step 5 is divided into rearward collision, frontal collisions, collision behind and static state and collides four kinds.
Adopt change speed in the collision prevention measure in the described step 5 or change two kinds of methods of direction and avoid collision, defined the condition of left collision prevention, right collision prevention, acceleration and four modules of deceleration and collision prevention thereof simultaneously.
In the described step 6 to collision prevention speed and direction is assessed and reorientation.
The invention has the beneficial effects as follows: in path planning, only the inside, walkable region of group is planned,, so just can reduce the algorithm number of times of collision prevention so there is no need to consider of the influence of static-obstacle thing to the pedestrian; This method is unit with the group, thus complexity with respect to the collision prevention of individuality, the quantity of collision detection obviously is much smaller; The method that this method is based on 2 detections is come carrying out collision prevention between the group, and this method is fairly simple in the prediction collision method, and in the individual case of collision of inner individual and other groups of consideration group, algorithm complex will further be simplified like this.
Description of drawings
Fig. 1 is the process flow diagram of the collision prevention method of a kind of multi-level virtual community of the present invention;
The simulation synoptic diagram of Fig. 2 for organizing among the present invention;
Fig. 3 is the rearward collision synoptic diagram of group among the present invention with group;
Fig. 4 is the frontal collisions synoptic diagram of group among the present invention with group;
Fig. 5 is the behind collision synoptic diagram of group among the present invention with group;
Fig. 6 is the static state collision synoptic diagram of group among the present invention with group.
Embodiment
Further specify below in conjunction with the preferred implementation of accompanying drawing to the collision prevention method of a kind of multi-level virtual community of the present invention, process flow diagram shown in Figure 1 has provided the process steps of realization of the present invention.In the method that this colony avoids colliding, mainly be by carrying out corresponding collision prediction respectively and avoid measure to finish collision prevention individual between colony accordingly to individual between group, the group.
In the present embodiment, the model format that environment loads is .dss, by the operation to environmental information, obtains its corresponding zone and current scene information and outputs in " map.txt ".Export current " indicateing arm " center in " positions.txt ".This method mainly is the collision prevention measure between group and the group, and its detailed process is:
Step 1 comes virtual community is divided into groups according to spatial division.Mainly be that position according to the static-obstacle thing is divided into walkable region and non-walkable region to environment space information, choose group according to the zone again, the individuality of the close together of same walkable region is divided into one group.The leader of the nearer individuality of distance center point as group.
Step 2, the simulation of group and the collision of organizing between the inner individuality are avoided.Because the behavior of group is similar with the flocks model, in order to simulate the group behavior, the individuality in the group should have identical speed, is ready to follow the leader together with the companion.A group must satisfy four kinds of behaviors: 1. collision prevention between the companion is avoided in collision; 2. speeds match keeps identical speed as far as possible; 3. cluster behavior is as far as possible near near companion; 4. the member of each group shares the IP tabulation from the initial point to the impact point.Provided the simulation drawing of organizing as Fig. 2, when the behavior of consideration group, there is a group radius in group, and individuality must remain within this radius with the leader.
The leader is exactly the agent of the consciousness of group, should organize the follower to pass than the big inlet of group radius simultaneously in seeking the footpath.The follower then is other individualities of following the leader, and generally speaking, the follower then keeps same speed with the leader as far as possible within the group radius, comprise size and Orientation.If the follower within the group radius, in order not bump with the agent (individuality) of barrier or other groups, must quicken to move towards leader's direction, after within the arrival group radius, carrying out speeds match.Promptly follow the leader that this behavior of leader makes a plurality of virtual pedestrian in the certain limit follow some appointments and advance, they show as possible near the leader, can not stop leader's road, and maintain a certain distance to avoid mutual collision with it.
Aspect the behavior of group, this method utilization control comes that inner individuality avoids colliding to organizing, and control by leader's attractive force and and adjacent partner between repulsive force etc. form.We are decomposed into cohesion to the behavior of group, separate and line up three behaviors, and then control also can be decomposed into the synthetic of three kinds of controls, and promptly at first, individuality all remains on same zone on the same group.If the individuality Wu Yuanliao that leaves one's post is arranged, should rejoin one's unit at once, cohesion behavior that Here it is, this behavior is by seeking all contiguous individualities of this individuality periphery, calculate their mean place or center, current location and this center of specified individual are subtracted each other, just can obtain ordering about individuality and be pooled to control in the group; Secondly, although individual, avoid can not running into mutually at same group.For this reason, it is too approaching that they all have a space to prevent with other individuality separately, promptly individual separating behavior, and the behavior obtains its control by the inverse that calculates the distance between two individualities, and individuality will remain on same direction simultaneously.Certainly angle separately is not necessarily identical, but general orientation is much the same, and this cries lines up, these three behaviors: condense, separate and line up, formed the behavior of group.
When the behavior of consideration group, just consider, and each individually just only need look at several partners of closing on, if by too near just from far point, gives me a little if the direction difference just turns over too much to organize as a whole going.Although splitted into several sub-behaviors on the behavioral techniques of group, yet, generally not too can only use wherein one or two behavior to the role always they almost occur simultaneously.
Step 3, the collision prediction between group and the group.This method is represented a group, P with A
AThe current position of expression A, S
ARepresent current speed, R
AIt then is the group radius (width) of group.Now suppose two adjacent group H and N, the expectation speed of H and N is respectively S
H, S
N, then the relative position of H and N and relative velocity are respectively: P
r=P
H-P
N, S
r=S
N-S
HSo the collision time between H and the N satisfies equation:
Wherein ε is the safe distance of H and N.If this equation is not separated or unique solution is arranged, then do not predict collision.If there are 2 to separate t
1, t
2, and t
1＜t
2, then: 1. t
2＜0: collision took place, so can not take place in the future again; 2. t
1＜0 and t
2＞0: collision takes place at once, need take measures to avoid the generation of colliding at once; 3. t
1〉=0: collision will be at t
1The back takes place.
Step 4 is determined the crash type between group and the group.If use t
cThe collision time of expression prediction, C
H=P
H+ S
H* t
c, C
N=P
N+ S
N* t
c, C wherein
HAnd C
NBe that H and N are at t
cPosition after time then has:
1. if (C
N-C
H) * S
H＜0, then be rearward collision (rear collision), as shown in Figure 3.
2. if (C
N-C
H) * S
H＞0 and S
H* S
N＜0, then be frontal collisions (front collision), as shown in Figure 4.
3. if (C
N-C
H) * S
H＞0 and S
H* S
N〉=0, then be collision (back collision) behind, as shown in Figure 5.
4. if || S
N||=0, then be static collision (static collision), as shown in Figure 6.
These four kinds of crash types propose for local collision prevention is set.
Step 5 is carried out collision prevention according to crash type.Can adopt two kinds of methods to avoid collision: change speed or change direction.Its method is to calculate the new speed S of group H
H', avoid bumping with group N.H has P with respect to N
r(t)=P
H-P
N+ (S
H-S
N) t.For fear of collision, then new relative velocity is S
r'=S
H'-S
NThis problem can be used S
H' another kind of geometric format represent i.e. straight line P
H-P
N+ S
r' * t be the center of circle at initial point, radius is R
H+ R
NThe tangent line of the round C of+ε.Suppose T
lWith T
rBe on the circle C and by tangent line P
rTwo points, D
l=T
l-(P
H-P
N), D
r=T
r-(P
H-P
N), when
And
The time, D
l, D
rBe to avoid the left and right directions that collides.D=D
lFor avoiding on a left side D=D
rFor avoiding on the right side.The α angle is rotated in R (α) expression, then by formula α D=S
H-S
N, and α＞0 is calculated collision and is avoided speed S
H'.Therefore, this method can define four kinds of methods and carries out collision prevention: left collision prevention (LA), right collision prevention (RA), quicken (A), (D) slows down.Each module will find new speed and new constraint condition:
(a) left and right collision prevention: || S
H' ||=|| S
H||, the size of maintenance speed.
(b) slow down: S
H'=β S
HAnd 0＜β＜1 keeps velocity reversal.
(c) quicken: S
H'=β S
HAnd β＞1 keeps velocity reversal.
Step 6, the assessment of collision prevention and reorientation.In considering colony's collision prevention process, the speed that produces for fear of the collision prevention between two groups then can be assessed the influence of guaranteeing other groups to collision prevention speed and direction to the influence of other groups.If speed S
HMay cause collision, and S
H' be the collision prevention speed of output, then Xia Mian rule will be used to assessment and reorientation:
n(S
H′，S
H，β
d，β
n)＝C
d(S
H′，S
H，β
d)*C
n(S
H′，S
H，β
n)
These three functions will calculate the mathematics factor between 0-1 and assess with respect to S
HSpeed S
H'.Function c
dThe difference of assessment direction, c
nThe difference of estimating velocity size.And the overall computing velocity S of n assessment
H' speed and direction, this function adopts two parameter beta
d, β
nCome the acute variation of equilibrium rate size and Orientation, and parameter beta
d, β
nAnd collision prevention measure and interrelated between the crash type (frontal collisions, rearward collision, collision behind, static collision) each time.See Table 1 navigation rule model and table 2 and minimize the regulation rule model, as β
d=1, β
n=1.In case after these two models are determined, these parameter informations will be used among the Local Optimization Algorithm.
Table 1 navigation rule model
| Measure type | Rearward collision | Frontal collisions | Collision behind | Static collision |
| Right collision prevention (RA) | n(S H′，S H，1，1) | ????n(S H′，S H，1，1) | ||
| Left side collision prevention (LA) | n(S H′，S H，1，1) | ????n(S H′，S H，1，1) | ||
| Quicken (A) | n(S H′，S H，1，1) | n(S H′，S H，1，1) | ||
| Slow down (D) | n(S H′，S H，1，5) | n(S H′，S H，1，5) |
Table 2 minimizes adjustment model
| Measure type | Rearward collision | Frontal collisions | Collision behind | Static collision |
| Right collision prevention (RA) | n(S H′，S H，1，1) | n(S H′，S H，1，1) | n(S H′，S H，1，1) |
| Measure type | Rearward collision | Frontal collisions | Collision behind | Static collision |
| Left side collision prevention (LA) | n(S H′，S H，1，1) | n(S H′，S H，1，1) | n(S H′，S H，1，1) | |
| Quicken (A) | n(S H′，S H，1，1) | n(S H′，S H，1，1) | ||
| Slow down (D) | n(S H′，S H，1，5) | n(S H′，S H，1，5) |
Step 7 is carried out local optimum according to assessment result, produces the collision prevention speed of the overall situation.Because the new collision prevention speed that calculates may produce new collision with other group, can new collision prevention speed and direction be optimized by Local Optimization Algorithm so, reduces the influence to other groups, produces a more excellent speed.
The collision time of supposing prediction is t
c, the time that another group bumps is t, the actual speed of group is v.
Situation is 1.: when predicting collision and prediction collision time t
c〉=t or basic just do not detect collision, then actual speed v will be chosen as the collision prevention speed of group.
Situation is 2.: when detecting collision and t
c＜t.If can confirm the crash type of this collision according to step 4, and the measure (left collision prevention, right collision prevention, acceleration, deceleration) of choosing a kind of collision prevention according to the method for step 5, then, choose the collision prevention speed of pairing speed as group according to the model of the table 2 in the step 6; Otherwise, forward step 3 to, up to the collision prevention speed of obtaining group.
Claims (5)
1. the collision prevention method of a multi-level virtual community is characterized in that, this method comprises:
Step 1 comes virtual community is divided into groups according to spatial division;
Step 2, the simulation of group and the collision of organizing between the inner individuality are avoided;
Step 3, the collision prediction between group and the group;
Step 4, determine the crash type between group and the group:
Step 5 is taked the collision prevention measure according to crash type;
Step 6, the assessment of collision prevention and reorientation;
Step 7 is carried out local optimum according to assessment result, produces the collision prevention speed of the overall situation.
2. the collision prevention method of a kind of multi-level virtual community according to claim 1, it is characterized in that, position according to the static-obstacle thing in the described step 1 is walkable region and non-walkable region with spatial division, in the certain distance scope in same walkable region Different Individual is divided into one group.
3. the collision prevention method of a kind of multi-level virtual community according to claim 1 is characterized in that, the crash type in described step 4 and the step 5 is divided into rearward collision, frontal collisions, collision behind and static state and collides four kinds.
4. the collision prevention method of a kind of multi-level virtual community according to claim 1, it is characterized in that, adopt change speed in the collision prevention measure in the described step 5 or change two kinds of methods of direction and avoid collision, defined the condition of left collision prevention, right collision prevention, acceleration and four modules of deceleration and collision prevention thereof simultaneously.
5. the collision prevention method of a kind of multi-level virtual community according to claim 1 is characterized in that, in the described step 6 to collision prevention speed and direction is assessed and reorientation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910265430 CN101719285B (en) | 2009-12-28 | 2009-12-28 | Collision prevention method for multi-layered virtual communities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910265430 CN101719285B (en) | 2009-12-28 | 2009-12-28 | Collision prevention method for multi-layered virtual communities |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101719285A true CN101719285A (en) | 2010-06-02 |
| CN101719285B CN101719285B (en) | 2013-01-02 |
Family
ID=42433854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910265430 Expired - Fee Related CN101719285B (en) | 2009-12-28 | 2009-12-28 | Collision prevention method for multi-layered virtual communities |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101719285B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012069022A1 (en) * | 2010-11-26 | 2012-05-31 | Technicolor (China) Technology Co., Ltd. | Method for animating characters, with collision avoidance based on tracing information |
| CN104281052A (en) * | 2013-07-06 | 2015-01-14 | 哈尔滨点石仿真科技有限公司 | Behavior based navigator-follower multi-agent formation control method |
| CN104299265A (en) * | 2014-10-22 | 2015-01-21 | 电子科技大学 | Group behavior control technology under fluid environment influence |
| CN104331917A (en) * | 2014-10-22 | 2015-02-04 | 电子科技大学 | Panic crowd escape simulation method |
| CN105550500A (en) * | 2015-12-08 | 2016-05-04 | 吉林大学 | Method for simulating following behavior of pedestrian based on social force |
| CN112527014A (en) * | 2020-12-02 | 2021-03-19 | 电子科技大学 | Unmanned aerial vehicle cluster grazing method based on packing algorithm |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2861857B1 (en) * | 2003-10-29 | 2006-01-20 | Snecma Moteurs | DISPLACEMENT OF A VIRTUAL ARTICULATED OBJECT IN A VIRTUAL ENVIRONMENT BY AVOIDING INTERNAL COLLISIONS BETWEEN THE ARTICULATED ELEMENTS OF THE ARTICULATED OBJECT |
| JP4383247B2 (en) * | 2004-05-14 | 2009-12-16 | 三菱プレシジョン株式会社 | Collision detection method and collision detection apparatus |
| JP2005342360A (en) * | 2004-06-07 | 2005-12-15 | Hitachi Ltd | Group character movement control method, recording medium, simulation device, and game device |
| CN100580708C (en) * | 2008-08-20 | 2010-01-13 | 浙江大学 | A Fuzzy Control Method for Generating Group Animation |
-
2009
- 2009-12-28 CN CN 200910265430 patent/CN101719285B/en not_active Expired - Fee Related
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012069022A1 (en) * | 2010-11-26 | 2012-05-31 | Technicolor (China) Technology Co., Ltd. | Method for animating characters, with collision avoidance based on tracing information |
| US9460540B2 (en) | 2010-11-26 | 2016-10-04 | Thomson Licensing | Method for animating characters, with collision avoidance based on tracing information |
| CN104281052A (en) * | 2013-07-06 | 2015-01-14 | 哈尔滨点石仿真科技有限公司 | Behavior based navigator-follower multi-agent formation control method |
| CN104299265A (en) * | 2014-10-22 | 2015-01-21 | 电子科技大学 | Group behavior control technology under fluid environment influence |
| CN104331917A (en) * | 2014-10-22 | 2015-02-04 | 电子科技大学 | Panic crowd escape simulation method |
| CN104299265B (en) * | 2014-10-22 | 2017-07-25 | 电子科技大学 | A Group Behavior Control Method Under the Influence of Fluid Environment |
| CN104331917B (en) * | 2014-10-22 | 2017-09-26 | 电子科技大学 | A kind of panic crowd's escape analogy method |
| CN105550500A (en) * | 2015-12-08 | 2016-05-04 | 吉林大学 | Method for simulating following behavior of pedestrian based on social force |
| CN105550500B (en) * | 2015-12-08 | 2017-12-15 | 吉林大学 | A kind of pedestrian's following behavior emulation mode based on social force |
| CN112527014A (en) * | 2020-12-02 | 2021-03-19 | 电子科技大学 | Unmanned aerial vehicle cluster grazing method based on packing algorithm |
| CN112527014B (en) * | 2020-12-02 | 2022-05-17 | 电子科技大学 | Unmanned aerial vehicle cluster grazing method based on packing algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101719285B (en) | 2013-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114384920B (en) | Dynamic obstacle avoidance method based on real-time construction of local grid map | |
| Essa et al. | Simulated traffic conflicts: do they accurately represent field-measured conflicts? | |
| Laugier et al. | Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety | |
| CN107730881A (en) | Traffic congestion vision detection system based on depth convolutional neural networks | |
| CN101719285B (en) | Collision prevention method for multi-layered virtual communities | |
| Zhu et al. | Interaction-aware cut-in trajectory prediction and risk assessment in mixed traffic | |
| CN107730903A (en) | Parking offense and the car vision detection system that casts anchor based on depth convolutional neural networks | |
| KR20130133778A (en) | Method for animating characters, with collision avoidance based on tracing information | |
| Joseph et al. | A novel hybrid deep learning algorithm for smart city traffic congestion predictions | |
| CN120257223B (en) | Highway vehicle track reconstruction method based on following model and dynamic time warping | |
| US20230314144A1 (en) | System and method for estimating drift path of marine floating body | |
| Sun et al. | Medium-fidelity evaluation and modeling for perception systems of intelligent and connected vehicles | |
| CN101431665A (en) | Method and system for detecting and tracking object | |
| CN111369053A (en) | A Trajectory Prediction Method for People and Vehicles in Human-Vehicle Interaction Scenarios | |
| Fu et al. | Trajectory Prediction and Risk Assessment in Car-Following Scenarios Using a Noise-Enhanced Generative Adversarial Network | |
| Chen et al. | The scanner of heterogeneous traffic flow in smart cities by an updating model of connected and automated vehicles | |
| Yang et al. | The research of multi-resolution modeling and simulation of the emergency evacuation | |
| CN118974707A (en) | Performance testing of robust systems | |
| Qiu et al. | Heterogeneous data‐based spatiotemporal trajectory synchronization for virtual–real interactive testing | |
| Cole et al. | A Sensor-Fused Deep Reinforcement Learning Framework for Multi-Agent Decision-Making in Urban Driving Environments | |
| Nasernejad | Modeling pedestrian behavior in pedestrian-vehicle near misses using inverse reinforcement learning algorithms | |
| Xu | Simulation and time-series analysis for Autonomous Emergency Braking systems | |
| CN117590865B (en) | Fixed wing unmanned aerial vehicle tracking target motion prediction method | |
| KR102844516B1 (en) | Outdoor Evacuation Simulation System for Disasters and Emergencies | |
| Velumail et al. | Federated vehicular tailgating detection utilizing distance and speed data analytics for intelligent transport system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130102 Termination date: 20171228 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |