CN106695791B - The continuous Free gait generation method of four-leg bionic robot - Google Patents
The continuous Free gait generation method of four-leg bionic robot Download PDFInfo
- Publication number
- CN106695791B CN106695791B CN201710017097.0A CN201710017097A CN106695791B CN 106695791 B CN106695791 B CN 106695791B CN 201710017097 A CN201710017097 A CN 201710017097A CN 106695791 B CN106695791 B CN 106695791B
- Authority
- CN
- China
- Prior art keywords
- foot
- robot
- point
- grid
- gravity
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000005021 gait Effects 0.000 title claims abstract description 49
- 239000011664 nicotinic acid Substances 0.000 title claims abstract description 18
- 230000005484 gravity Effects 0.000 claims abstract description 73
- 238000011084 recovery Methods 0.000 claims abstract description 6
- 230000004308 accommodation Effects 0.000 claims abstract description 5
- 210000002683 foot Anatomy 0.000 claims description 116
- 230000033001 locomotion Effects 0.000 claims description 36
- 238000011156 evaluation Methods 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 239000004576 sand Substances 0.000 claims description 3
- 238000012876 topography Methods 0.000 abstract description 7
- 238000013439 planning Methods 0.000 description 8
- 238000005265 energy consumption Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1661—Programme controls characterised by programming, planning systems for manipulators characterised by task planning, object-oriented languages
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1669—Programme controls characterised by programming, planning systems for manipulators characterised by special application, e.g. multi-arm co-operation, assembly, grasping
Abstract
A kind of continuous Free gait generation method of four-leg bionic robot, comprising the following steps: (1) obtain landform point cloud data, obtain the elevation map of landform, compare the threshold value preestablished, judge that sufficient region can not be fallen in landform；(2) when each foot of calculating robot is chosen to be swing foot, guarantee robot centre of gravity adjustment amount required for sufficient stability recovery phase；(3) it according to each sufficient position of robot, selectes and swings foot；(4) robot is according to the centre of gravity adjustment amount active accommodation center of gravity of calculating to target position；(5) location information of Combining with terrain information and robot selects the corresponding position point in landform to fall foot point as optimal to swing foot；(6) swing foot be rocked to it is optimal fall foot point, when the vola for swinging foot bottoms out sensor and detects and contacts to earth, go to step (2).This method can make robot avoid the occurrence of the case where causing unstability due to because falling being enough to that sufficient region can not be fallen, thus independently, stablize, quickly move through complicated rugged topography.
Description
Technical field
The present invention relates to a kind of method for generating the continuous Free gait of four-leg bionic robot, using this method, four
Sufficient bio-robot can belong to Robot Control Technology quickly and stably by the inclusion of the complicated landform that can not fall sufficient region
Field.
Background technique
Quadruped robot can be in the rugged higher physical relief of degree, more even for wheeled and crawler frame
Stabilized walking in the complicated landform environment that people can not cross over.Often there is be unsuitable for four-footed in complicated physical relief environment
Robot falls sufficient region, and when robot, which is fallen, is enough to these regions, easily it is steady to influence robot for generation vola sliding, depression etc.
Qualitative situation.Thus, quadruped robot just allows for accurately obtaining when walking in the landform comprising that can not fall sufficient region
Terrain information, and then by handling accordingly, obtain in landform be unsuitable for robot fall it is sufficient can not fall sufficient region, and be based on
This, proposes corresponding gait planning method.
Quadruped robot fall foot point be it is discrete, to guarantee robot stabilizedly by the inclusion of the rugged of sufficient region can not be fallen
Landform, and avoid falling the sufficient region that can not fall being enough in landform, quadruped robot must use the quiet gait of aperiodicity ---
Free gait.
In the Related Research Domain of quadruped robot, Free gait planning is to improve quadruped robot terrain adaptability
Key factor.Currently, domestic and international researchers have been presented for the generation method of a variety of Free gaits.1999 " Robotica "
(" robot ") is in 17 (4): paper " the Quadruped free gait generation based on that 405-412 is delivered
The primary/secondary gait " in, propose a kind of entitled " main/secondary " Free gait generation method, machine
Device people in the process of walking, is mainly walked with periodically quiet gait, only when not being available periodically quiet gait walking, is just turned
And it is walked forward using Free gait；2002, " The International Journal of Robotics
Research " (" international robot research magazine ") in 21 (2): paper " the Free gaits for that 115-130 is delivered
Quadruped robots over irregular terrain " propose three kinds of different types of Free gaits: free crab row
Gait, free rotation gait and gait is freely turned to, robot can be comprising can not the area Luo Zu using these three Free gaits
It is upper in any direction in the landform in domain to advance；2015, in " planning of quadruped robot obstacle detouring Free gait is studied with control "
Propose a kind of method that plain algorithm generation Free gait is searched based on figure.In these gait plannings, in the prevalence of not providing
Specifically to landform Information acquisition the problem of, and accurately handle terrain information is that quadruped robot is made to avoid falling and be enough to
The premise in sufficient region can not be fallen in landform.In addition, the center of gravity of robot is only in direction of advance in the planning of these Free gaits
Mobile, stability is not high, limits the terrain adaptability of robot.Chinese patent literature CN104267720A discloses " a kind of
The Free gait generation method of four-leg bionic robot ", a kind of discontinuous Free gait is produced using the gait generation method,
When robot is walked with this gait, there is the lower deficiencies of average movement velocity.The gait generation method is to be based on mentioning
Optional in preceding given landform falls what foot point was planned, and foot point can not be fallen by not providing to be analyzed in landform according to landform information independence
The algorithm in region is not high with practical application request compactness.In addition, in Free gait planning, the determination method for sequence of taking a step
It is affected to robot motion's performance, and in the gait generation method, it is only true using kinematics nargin as sequence of taking a step
Fixed foundation does not comprehensively consider the other factors for influencing robot motion's performance.
In view of this, to improve independence of the quadruped robot when walking in the landform comprising sufficient region can not be fallen, it must
Processing terrain information must be given by and obtain the processing method that can not fall sufficient region in landform；Meanwhile for four-leg bionic machine
People's Free gait generate in take a step sequence determination, gravity motion planning etc., four-leg bionic robot need to be comprehensively considered and existed
Stability, rapidity and flexibility in walking process are generated using the terrain adaptability for improving four-leg bionic robot as target
Meet the Free gait of actual demand.
Summary of the invention
It is imitative to provide a kind of four-footed for many deficiencies existing for existing quadruped robot Free gait generation method by the present invention
The raw continuous Free gait generation method of robot.This method is answered based on providing to accurately identifying for terrain information with processing
With this method, robot can be made to avoid the occurrence of the case where causing unstability due to because falling being enough to that sufficient region can not be fallen, thus independently, surely
Rugged topography that is fixed, quicklying move through complexity.
Continuous Free gait planing method provided by the invention is based on providing to accurately identifying for terrain information with processing
's.By the processing to terrain information, four-leg bionic robot, which can specify in its own position and landform, can not fall sufficient region
Relative positional relationship, fall foot point Algorithms of Selecting in conjunction with optimal, can effectively guarantee that robot avoids falling and be enough to that foot can not be fallen
Region；To guarantee that robot has enough stability margins in the process of walking, robot passes through the center of gravity of four-footed driving phase
It swings, to increase the stability margin of robot；Robot takes a step sequentially to be unfixed when walking using Free gait,
In four-footed support phase, comprehensively consider corresponding heavy when the projection of robot center of gravity is swung with the relationship of support polygon and each foot
Heart amount of movement can be protected with determining the sequence of taking a step of robot in the process of walking while guaranteeing robot flexible motion
Card robot maintains the stability of its own with the smallest amount of movement；In addition, to improve the average movement velocity of quadruped robot,
The trunk of planning robot is constantly travelled forward along direction of advance with given speed.
The continuous Free gait generation method of four-leg bionic robot of the invention, specifically includes the following steps:
(1) landform point cloud data is obtained, simplified and rasterizing is carried out to point cloud data, and obtain the elevation map of landform；Root
According to the elevation map of landform, the corresponding variance of each grid is calculated, the threshold value preestablished is compared, judges that foot can not be fallen in landform
Region；
(2) it is required to guarantee stability of the robot in sufficient recovery phase when each foot of calculating robot is chosen to be swing foot
The centre of gravity adjustment amount wanted；
(3) according to each sufficient position of robot, with reduce robot gravity motion number and center of gravity in side direction
Amount of movement is principle, selectes and swings foot；
(4) robot is according to the centre of gravity adjustment amount active accommodation center of gravity of calculating to target position；
(5) location information of Combining with terrain information and robot, according to the optimal foot point selection algorithm that falls for swing foot selection
Corresponding position point in landform falls foot point as optimal；
(6) swing foot be rocked to it is optimal fall foot point, when the vola for swinging foot bottoms out sensor and detects and contacts to earth, go to step
(2), whole process is repeated.
Carrying out simplified and rasterizing process to point cloud data in the step (1) is:
1. determining the size of single grid, length × width is m × n (unit mm)；
2. point cloud data is carried out rasterizing processing according to determining lattice dimensions；
3. finding out the corresponding fit Plane of each grid middle data respectively；
4. the height value with the height value of fit Plane corresponding to grid midpoint, as the grid.
Judge that the process that can not fall sufficient region in landform is in the step (2):
By establishing assessment models, to assess the rugged degree of each grid, assess some grid it is rugged spend when, together with
(P*Q-1) a grid collectively constitutes assessment models around this grid and grid, wherein P and Q is odd number；
The height value that grid is indicated with E (i, j), then acquire grid in the assessment models of the rugged degree of grid (i, j) respectively
Dispersed elevation value and height value variance, are shown below:
Rugged degree evaluation of estimate using the variance of grid (i, j) assessment models as grid, is calculated the rugged of each grid
After rugged degree evaluation of estimate, compared with preset threshold value T-phase；If the rugged degree evaluation of estimate of some grid meets following formula, recognize
It is that can not fall sufficient region for the region where the grid:
S(i,j)2≥T。
The determination principle that swing foot is selected in the step (3) is:
(1) if you do not need to gravity motion, certain only foot when swinging, robot can meet minimum stability margin constraint with
And when kinematic constraint, then this can be preferred completely to swing foot；
(2) if all foots do not meet (1) article, the centre of gravity adjustment amount according to calculating is needed, obtains four foots point
When not being selected as swinging sufficient, corresponding centre of gravity adjustment amount: Ly1、Ly2、Ly3And Ly4；Compare this four centre of gravity adjustment amounts later
Size selects the minimum corresponding foot of gravity motion amount sufficient as swinging.
Centre of gravity adjustment amount calculating process in the step (4) is:
When four-footed supports the adjusting stage, calculates the adjustment amount of quadruped robot center of gravity, it is necessary to follow following two rule:
(1) guarantee that quadruped robot after centre of gravity adjustment, is swung in sufficient swing process, stability margin is not less than preparatory
The minimum stability margin S of settingmin；
(2) while meeting regular (1), the centre of gravity adjustment amount in side direction is minimized；
Assuming that the movement velocity in Robot direction of advance is Vx, the time in centre of gravity adjustment stage is t1, then center of gravity tune is obtained
Whole source location P1Abscissa x1Following formula need to be met:
xI=Lx1=Vx·t1,
Quadruped robot to be made is after the centre of gravity adjustment of four-footed driving phase, and the stability margin of quadruped robot is not less than most
Small stability margin Smin, then point P1Position need to meet following formula:
(Lx1+Lx2)≥Smin,
If the time in centre of gravity adjustment stage is t2, then Lx2Value acquired by following formula:
Lx2=Vx·t2,
Quadruped robot to be made is after the recovery phase of foot, and quadruped robot still ensures that the stability of its own, i.e., surely
Determine nargin and is not less than minimum stability margin Smin, then after the stage of taking a step, the source location P of center of gravity projection2Under need to meeting
Formula:
Lx3≥Smin,
Wherein, Lx3For point P2Along direction of advance to the cornerwise linear distance of support.
To improve the energy consumption of robot in the process of walking, point P1With point P2Meet formula (Lx1+Lx2)≥SminAnd Lx3≥Smin
When, corresponding minimum movement amount Ly, the as centre of gravity adjustment amount of target.
Optimal in the step (5) falls foot point selection algorithm, is:
Wherein, cjWhat is represented is to swing j-th of foot optionally to fall foot point；WithWhat is respectively indicated is optional to fall for j-th
Transverse and longitudinal coordinate of the foot point in world coordinate system { W }；
It swings the optimal of foot and falls foot point, it is necessary to meet following two condition:
(1) it is optimal fall foot point and must help to swing foot obtain step-length big as far as possible, facilitating robot in this way, to improve it flat
Equal movement velocity；
(2) the optimal foot point that falls must help robot to obtain greater area of stability region, facilitate robot in this way and mention
Its high stability during the motion；
With Li(x, y) and Si(x, y), which is respectively indicated, optionally to be fallen foot point for i-th and is confirmed as optimal when falling foot point, and foot is swung
Step-length and the area of corresponding support triangle shape then establish the optimal evaluation function for falling foot point, be shown below:
F (x)=wS·Li(x,y)+wL·Si(x, y),
Wherein, wSAnd wLWhat is respectively indicated is the weight coefficient for swinging sufficient step-length and robot support triangle shape area；
According to the optional coordinate for falling foot point stored in array C, calculates separately out and each optional fall the corresponding evaluation of foot point
The value of function, wherein the corresponding optional foot point that falls of the value of maximum evaluation function is to swing the optimal of foot to fall foot point.
The invention has the characteristics that:
1. combining the landform point cloud information obtained, four-leg bionic robot, which can be identified accurately in landform, can not fall foot
Region；
2. can help Robot Selection is optimal to fall foot point, and then guarantee that robot can not fall sufficient region in avoiding landform
While, increase robot stabilization and average movement velocity；
3. effectively increasing the stability margin of quadruped robot in the process of walking, in turn by the active accommodation of center of gravity
The terrain adaptability of quadruped robot can be improved；
4. optimizing sequence of taking a step determines that method reduces motion process while ensuring robot motion's flexibility
In lateral adjustment amount, the energy consumption of robot can be effectively reduced；
5. robot is during the motion, trunk is constantly travelled forward in direction of advance with preset speed, is had
The movement velocity for improving to effect robot, makes robot under the premise of stabilization, passes through complicated landform in the shortest possible time.
Detailed description of the invention
Fig. 1 is the simulation model figure of ten two degrees of freedom four-leg bionic robots.
Fig. 2 is the point cloud data figure of the rugged topography obtained by TOF camera.
Fig. 3 is the point cloud data figure for including in grid (i, j).
Fig. 4 is the fit Plane schematic diagram of point cloud data in grid (i, j).
Fig. 5 is the elevation map of rugged topography after simplifying.
Fig. 6 is the rugged degree evaluation model figure of grid.
Fig. 7 is that sufficient area schematic can not be fallen in rugged topography.
Fig. 8, which is that quadruped robot is optimal, falls foot point selection exemplary diagram.
Fig. 9 is that robot centre of gravity adjustment amount calculates exemplary diagram.
Figure 10 is to swing sufficient selection method exemplary diagram.
Specific embodiment
It is continuously free to quadruped robot of the invention below by taking ten two degrees of freedom quadruped robot shown in FIG. 1 as an example
The generation method of gait is described in detail.
The processing of one, terrain information
Firstly, obtaining the point cloud number of landform by light transmission time (Time Of Flight, TOF) three-dimensional laser camera
According to as shown in Figure 2.Position of each point in world coordinate system { W } in cloud will be put, will be stored into digital P, as shown in following formula (1).
Wherein, pi(xi,yi,zi) what is indicated is the i-th point of coordinate in world coordinate system in a cloud.
In the processing of terrain information, it is divided into two steps, the first step is the simplification to terrain data；Second step is over the ground
The assessment of the rugged degree of shape.
(1) terrain data simplifies
In order to reduce point cloud data, to reduce calculation amount, following processing is done to point cloud data:
1. determining the size of single grid, m × n (length × wide, unit mm)；
2. point cloud data is carried out rasterizing processing according to determining lattice dimensions；
3. finding out the corresponding fit Plane of each grid middle data respectively；
4. the height value with the height value of fit Plane corresponding to grid midpoint, as the grid.
Using grid (i, j) shown in Fig. 3 and point cloud data wherein included as example, illustrate the solution of fit Plane
Process.
Firstly, the equation of fit Plane is provided, as shown in following formula (2).
Z=Ax+By+C (2)
Wherein, A, B, C are undetermined coefficient.
It according to formula (2), can be obtained in grid, the sum of the distance of each point to fit Plane, as shown in following formula (3).
Wherein, what m was indicated is the sum of point data in grid (i, j).
According to formula (3), formula (3) can be obtained to the partial derivative of each parameter, as shown in formula (4).
For the fit Plane for acquiring each point in grid, the equation group as shown in following formula (5) is constructed.By acquiring in formula (5)
Parameter in formula (2) can be obtained in equation group.
According to formula (2), formula (4) and formula (5), the matrix equation as shown in formula (6) can be obtained.
In turn, according to formula (6), the matrix equation of each parameter in solution formula (2) can be obtained, as shown in following formula (7).
After acquiring parameter A, B and C, the fit Plane of point cloud data in grid (i, j) shown in Fig. 3, such as Fig. 4 can be obtained
It is shown.
It is acquired in formula (2) after each coefficient by formula (7), the height value of grid (i, j) can be obtained.
The method provided according to this part obtains height as shown in Figure 5 after Point Cloud Processing shown in Fig. 2
Cheng Tu.
(2) inequality is assessed
By establishing assessment models, to assess the rugged degree of each grid.Assess some grid it is rugged spend when, together with
(P*Q-1) a grid collectively constitutes assessment models around this grid and grid, wherein P and Q is odd number.Fig. 6 gives
Assess the assessment models example of grid (i, j) rugged degree, wherein P=5, Q=3.
The height value that grid is indicated with E (i, j), then can acquire grid in the assessment models of the rugged degree of grid (i, j) respectively
Dispersed elevation value and height value variance, as shown in following formula (9).
Rugged degree evaluation of estimate using the variance of grid (i, j) assessment models as grid, is calculated the rugged of each grid
After rugged degree evaluation of estimate, compared with preset threshold value (T).If the rugged degree evaluation of estimate of some grid meets formula (10),
Region where then thinking the grid is that can not fall sufficient region.
S(i,j)2≥T (10)
The example provided with Fig. 7 can not fall sufficient region using what the above method obtained landform, as white area is in Fig. 7
Quadruped robot can not fall sufficient region in rugged topography.
Two, are optimal to fall foot point selection algorithm
Foot point selection example is fallen in conjunction with what Fig. 8 was provided, provides and optimal falls foot point selection algorithm.Solid box shown in fig. 8
For the effective working region for swinging foot, the crosspoint of transverse and longitudinal dotted line is optionally to fall foot point, and the region of shade covering is that can not fall
Sufficient region.The optional coordinate position for falling foot point is stored in array C, as shown in following formula (11).
Wherein, cjWhat is represented is to swing j-th of foot optionally to fall foot point；WithWhat is respectively indicated is optional to fall for j-th
Transverse and longitudinal coordinate of the foot point in world coordinate system { W }.
It swings the optimal of foot and falls foot point, it is necessary to meet following two condition:
(1) it is optimal fall foot point and must help to swing foot obtain step-length big as far as possible, facilitating robot in this way, to improve it flat
Equal movement velocity；
(2) the optimal foot point that falls must help robot to obtain greater area of stability region, facilitate robot in this way and mention
Its high stability during the motion.
With Li(x, y) and Si(x, y), which is respectively indicated, optionally to be fallen foot point for i-th and is confirmed as optimal when falling foot point, and foot is swung
Step-length and the area of corresponding support triangle shape then establish the optimal evaluation function for falling foot point, as shown in formula (12).
F (x)=wS·Li(x,y)+wL·Si(x,y) (12)
Wherein, wSAnd wLWhat is respectively indicated is the weight coefficient for swinging sufficient step-length and robot support triangle shape area.
According to the optional coordinate for falling foot point stored in array C, it can calculate separately out and each optional fall that foot point is corresponding to be commented
The value of valence function, wherein the corresponding optional foot point that falls of the value of maximum evaluation function is to swing the optimal of foot to fall foot point.
Three, centre of gravity adjustment amounts calculate
In the quadruped robot Free gait planing method that the present invention provides, robot is measured using longitudinal stability nargin
Stability.
In four-footed driving phase, robot completes the active accommodation of center of gravity, it is ensured that swinging foot can steadily swing
Foot point is fallen to optimal.But swing of the center of gravity in direction of advance side direction, also improve the energy consumption of quadruped robot.
Thus, the adjusting stage is supported in four-footed, when calculating the adjustment amount of quadruped robot center of gravity, it is necessary to follow following two
Rule:
(1) guarantee that quadruped robot after centre of gravity adjustment, is swung in sufficient swing process, stability margin is not less than preparatory
Minimum stability margin (the S of settingmin)；
(2) while meeting regular (1), the centre of gravity adjustment amount in side direction is minimized.
Assuming that the movement velocity in Robot direction of advance is Vx, the time in centre of gravity adjustment stage is t1, then Fig. 9 can be obtained
In, the source location P of centre of gravity adjustment1Abscissa x1Following formula (13) need to be met.
xI=Lx1=Vx·t1 (13)
Quadruped robot to be made is after the centre of gravity adjustment of four-footed driving phase, and the stability margin of quadruped robot is not less than most
Small stability margin Smin, then point P1Position need to meet following formula (14).
(Lx1+Lx2)≥Smin (14)
If the time in centre of gravity adjustment stage is t2, then Lx2Value can be acquired by formula (15).
Lx2=Vx·t2 (15)
Quadruped robot to be made is after the recovery phase of foot, and quadruped robot still ensures that the stability of its own, i.e., surely
Nargin is determined not less than Smin, then after the stage of taking a step, the source location P of center of gravity projection2Formula (16) need to be met.
Lx3≥Smin (16)
Wherein, Lx3For point P2Along direction of advance to the cornerwise linear distance of support.
To improve the energy consumption of robot in the process of walking, point P1With point P2When meeting formula (14), formula (16), it is corresponding most
Small amount of movement Ly, the as centre of gravity adjustment amount of target.
Four, take a step the determination of sequence
For quadruped robot when being walked forward using Free gait, sequence of taking a step is unfixed.Thus, four-footed machine
People needs independently to determine its sequence of taking a step in the process of walking.
It, be to reduce quadruped robot when sequence is taken a step in determination in the Free gait generation method provided in the present invention
Amount of movement of the number and center of gravity of gravity motion in side direction is principle, so as to improve the capacity usage ratio of quadruped robot.
With specific example shown in Fig. 10, the determination principle for swinging foot is introduced.
(1) if you do not need to gravity motion, certain only foot when swinging, robot can meet minimum stability margin constraint with
And when kinematic constraint, then this can be preferred completely to swing foot.In this way, machine reduces the mobile number of center of gravity, it can be effective
Ground reduces energy consumption.
By taking such as Figure 10 as an example, when being determined as No. 1 foot to swing sufficient, the projection of robot center of gravity needs not move through lateral
Adjustment that is, in support triangle shape, and meets formula (17).
L≥Smin (17)
In addition, the value of L meets formula (18) in Figure 10.
L≥[Vx·(t1+t2)+Smin] (18)
It follows that the projection of robot center of gravity needs not move through lateral tune when being determined as No. 1 foot to swing sufficient
It is whole, stable constraint (formula (17)) and kinematical constraint (formula (18)) can be met.Thus, by No. 1 foot as foot is swung, in this way
The adjustment campaign an of center of gravity can be saved, and then energy consumption can be lowered.
(2) it if all foots are not inconsistent normally (1), needs according to the center of gravity provided in " calculating of 4 centre of gravity adjustment amounts "
Adjustment amount calculation method, when obtaining four foots and being selected as swinging sufficient respectively, corresponding centre of gravity adjustment amount: Ly1、Ly2、Ly3And Ly4。
The size for comparing this four centre of gravity adjustment amounts later selects the minimum corresponding foot of gravity motion amount sufficient as swinging.
According to above-mentioned two rule, quadruped robot automatically can determine that the swing for swinging foot is suitable in the process of walking
Sequence, and ultimately form the sequence of taking a step suitable for quadruped robot when walking in sufficient region rugged topography comprising that can not fall.
Claims (6)
1. a kind of continuous Free gait generation method of four-leg bionic robot, characterized in that the following steps are included:
(1) landform point cloud data is obtained, and carries out simplified and rasterizing, obtains the elevation map of landform；According to the elevation map of landform,
The corresponding variance of each grid is calculated, the threshold value preestablished is compared, judges that sufficient region can not be fallen in landform；
(2) it is required to guarantee stability of the robot in sufficient recovery phase when each foot of calculating robot is chosen to be swing foot
Centre of gravity adjustment amount；
(3) according to each sufficient position of robot, with reduce quadruped robot gravity motion number and center of gravity in side direction
Amount of movement is principle, selectes and swings foot；
(4) robot is according to the centre of gravity adjustment amount active accommodation center of gravity of calculating to target position；
(5) location information of Combining with terrain information and robot, according to the optimal foot point selection algorithm that falls for swing foot selection landform
In corresponding position point fall foot point as optimal；
(6) swing foot be rocked to it is optimal fall foot point, when the vola for swinging foot bottoms out sensor and detects and contacts to earth, go to step (2),
Repeat whole process.
2. the continuous Free gait generation method of four-leg bionic robot according to claim 1, characterized in that the step
(1) carrying out simplified and rasterizing process to point cloud data in is:
1. determining the size of single grid, length × width is m × n；
2. point cloud data is carried out rasterizing processing according to determining lattice dimensions；
3. finding out the corresponding fit Plane of each grid middle data respectively；
4. the height value with the height value of fit Plane corresponding to grid midpoint, as the grid.
3. the continuous Free gait generation method of four-leg bionic robot according to claim 1, characterized in that the step
(1) judge that the process that can not fall sufficient region in landform is in:
By establishing assessment models, to assess the rugged degree of each grid, assess some grid it is rugged spend when, together with this
(P*Q-1) a grid collectively constitutes assessment models around grid and grid, wherein P and Q is odd number；
The height value that grid is indicated with E (i, j) then acquires being averaged for grid in the assessment models of the rugged degree of grid (i, j) respectively
Height value and height value variance, are shown below:
Rugged degree evaluation of estimate using the variance of grid (i, j) assessment models as grid, is calculated the rugged degree of each grid
After evaluation of estimate, compared with preset threshold value T-phase；If the rugged degree evaluation of estimate of some grid meets following formula, then it is assumed that should
Region where grid is that can not fall sufficient region:
S(i,j)2≥T。
4. the continuous Free gait generation method of four-leg bionic robot according to claim 1, characterized in that the step
(3) the determination principle that swing foot is selected in is:
(1) if you do not need to gravity motion, only for foot when swinging, robot can meet minimum stability margin constraint and fortune for certain
When moving constraint, then this can be preferred completely to swing foot；
(2) if all foots do not meet (1) article, the centre of gravity adjustment amount according to calculating is needed, four foots is obtained and selects respectively
When to swing foot, corresponding centre of gravity adjustment amount: Ly1、Ly2、Ly3And Ly4；Compare the size of this four centre of gravity adjustment amounts later,
Select the minimum corresponding foot of gravity motion amount sufficient as swinging.
5. the continuous Free gait generation method of four-leg bionic robot according to claim 1, characterized in that the step
(4) the centre of gravity adjustment amount calculating process in is:
When four-footed supports the adjusting stage, calculates the adjustment amount of quadruped robot center of gravity, it is necessary to follow following two rule:
(1) guarantee that quadruped robot after centre of gravity adjustment, is swung in sufficient swing process, stability margin, which is not less than, to be preset
Minimum stability margin Smin；
(2) while meeting regular (1), the centre of gravity adjustment amount in side direction is minimized；
Assuming that the movement velocity in Robot direction of advance is Vx, the time in centre of gravity adjustment stage is t1, then centre of gravity adjustment is obtained
Source location P1Abscissa x1Following formula need to be met:
xI=Lx1=Vx·t1,
For quadruped robot to be made after the centre of gravity adjustment of four-footed driving phase, the stability margin of quadruped robot is steady not less than minimum
Determine nargin Smin, then point P1Position need to meet following formula:
(Lx1+Lx2)≥Smin,
If the time in centre of gravity adjustment stage is t2, then Lx2Value acquired by following formula:
Lx2=Vx·t2,
For quadruped robot to be made after the recovery phase of foot, quadruped robot still ensures that the stability of its own, that is, stablizes abundant
Degree is not less than minimum stability margin Smin, then after the stage of taking a step, the source location P of center of gravity projection2Following formula need to be met:
Lx3≥Smin,
Wherein, Lx3For point P2Along direction of advance to supporting cornerwise linear distance,
Point P1With point P2Meet formula (Lx1+Lx2)≥SminAnd Lx3≥SminWhen, corresponding minimum movement amount Ly, the as center of gravity of target
Adjustment amount.
6. the continuous Free gait generation method of four-leg bionic robot according to claim 1, characterized in that the step
(5) optimal in falls foot point selection algorithm, is:
The optional coordinate position for falling foot point is stored in array C, is shown below:
Wherein, cjWhat is represented is to swing j-th of foot optionally to fall foot point；CxjWithCyjWhat is respectively indicated is optional to fall foot point j-th
Transverse and longitudinal coordinate in world coordinate system { W }；
It swings the optimal of foot and falls foot point, it is necessary to meet following two condition:
(1) the optimal foot point that falls must help to swing foot acquisition step-length big as far as possible, facilitate robot in this way and improve its average fortune
Dynamic speed；
(2) the optimal foot point that falls must help robot to obtain greater area of stability region, facilitate robot in this way and improve it
Stability during the motion；
With Li(x, y) and Si(x, y), which is respectively indicated, optionally to be fallen foot point for i-th and is confirmed as optimal when falling foot point, and sufficient step is swung
Then the area of long and corresponding support triangle shape establishes the optimal evaluation function for falling foot point, is shown below:
F (x)=wS·Li(x,y)+wL·Si(x, y),
Wherein, wSAnd wLWhat is respectively indicated is the weight coefficient for swinging sufficient step-length and robot support triangle shape area；
According to the optional coordinate for falling foot point stored in array C, calculates separately out and each optional fall the corresponding evaluation function of foot point
Value, wherein the value of maximum evaluation function it is corresponding it is optional fall foot point be swing foot optimal fall foot point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710017097.0A CN106695791B (en) | 2017-01-11 | 2017-01-11 | The continuous Free gait generation method of four-leg bionic robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710017097.0A CN106695791B (en) | 2017-01-11 | 2017-01-11 | The continuous Free gait generation method of four-leg bionic robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106695791A CN106695791A (en) | 2017-05-24 |
| CN106695791B true CN106695791B (en) | 2019-02-05 |
Family
ID=58907193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710017097.0A Expired - Fee Related CN106695791B (en) | 2017-01-11 | 2017-01-11 | The continuous Free gait generation method of four-leg bionic robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106695791B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107538490A (en) * | 2017-08-25 | 2018-01-05 | 哈尔滨理工大学 | Towards the quadruped robot motion planning method of complicated landform |
| CN109696909B (en) * | 2017-10-23 | 2022-04-15 | 深圳市优必选科技有限公司 | Foot type robot path planning method and device |
| CN107807655B (en) * | 2017-10-27 | 2019-08-30 | 山东大学 | Quadruped robot balance and stability control method based on gait adjustment in three-dimensional space |
| CN109753982B (en) * | 2017-11-07 | 2021-09-03 | 北京京东乾石科技有限公司 | Obstacle point detection method, obstacle point detection device, and computer-readable storage medium |
| CN107914271A (en) * | 2017-11-15 | 2018-04-17 | 上海与德科技有限公司 | A kind of control method and control system based on robot |
| CN108333931B (en) * | 2018-01-25 | 2020-10-02 | 北京理工大学 | Rugged terrain-oriented four-legged robot double-layer structure gait planning method |
| CN110262532B (en) * | 2019-06-24 | 2020-11-27 | 重庆大学 | Robot terrain processing and multi-terrain gait control method and system |
| CN110328670B (en) * | 2019-08-27 | 2022-05-27 | 山东科技大学 | Terrain fuzzy self-adaptive four-footed robot static gait planning method |
| CN110502857B (en) * | 2019-08-27 | 2022-12-27 | 山东科技大学 | Terrain roughness online estimation method for quadruped robot |
| CN110497414B (en) * | 2019-08-27 | 2022-04-26 | 山东科技大学 | Method for planning continuous body motion trail of quadruped robot |
| CN110815211B (en) * | 2019-09-29 | 2021-04-23 | 中国北方车辆研究所 | Method for quadruped robot to dynamically cross convex obstacle |
| CN111288985A (en) * | 2020-03-04 | 2020-06-16 | 北京易控智驾科技有限公司 | Map determination method and device, equipment and automatic mine car driving method |
| CN111766885B (en) * | 2020-07-08 | 2023-08-11 | 清华大学深圳国际研究生院 | Static gait planning method of quadruped robot |
| CN114489092B (en) * | 2020-10-26 | 2023-07-18 | 腾讯科技（深圳）有限公司 | Foot robot motion control method, device, equipment and medium |
| CN112697149B (en) * | 2020-12-10 | 2022-08-16 | 华南理工大学 | Planning method for rhythm gait foot end track of hexapod robot |
| CN112947312B (en) * | 2021-01-28 | 2021-09-14 | 仲恺农业工程学院 | Agricultural robot motion control method |
| CN113071577B (en) * | 2021-04-30 | 2022-10-25 | 郑州宇通重工有限公司 | Robot sole force flexible control method and multi-foot robot platform |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1486299B1 (en) * | 2002-03-15 | 2012-03-14 | Sony Corporation | Operation control device for leg-type mobile robot and operation control method, and robot device |
| US8457830B2 (en) * | 2010-03-22 | 2013-06-04 | John R. Goulding | In-line legged robot vehicle and method for operating |
| CN103192898A (en) * | 2013-04-09 | 2013-07-10 | 北京交通大学 | Biomimetic quadruped robot provided with head and tail balance adjustment devices |
| CN104267720B (en) * | 2014-10-24 | 2017-01-11 | 山东大学 | Method for generating free gaits for four-footed bionic robot |
| CN104986241B (en) * | 2015-06-29 | 2018-04-24 | 山东大学（威海） | The gait planning method of quadruped robot |
-
2017
- 2017-01-11 CN CN201710017097.0A patent/CN106695791B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN106695791A (en) | 2017-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106695791B (en) | The continuous Free gait generation method of four-leg bionic robot | |
| CN108333931B (en) | Rugged terrain-oriented four-legged robot double-layer structure gait planning method | |
| CN111913490B (en) | Four-foot robot dynamic gait stability control method and system based on foot falling adjustment | |
| CN110328670A (en) | The quiet gait planning method of quadruped robot based on landform fuzzy self-adaption | |
| Englsberger et al. | Three-dimensional bipedal walking control using divergent component of motion | |
| CN111766885B (en) | Static gait planning method of quadruped robot | |
| CN104062902B (en) | Delta robot time optimal trajectory planning method | |
| CN103085070B (en) | Quadruped robot motion planning method for facing complex terrain | |
| CN104267720B (en) | Method for generating free gaits for four-footed bionic robot | |
| CN109093626A (en) | The fuselage attitude control method and device of quadruped robot | |
| CN105354875A (en) | Construction method and system for two-dimensional and three-dimensional joint model of indoor environment | |
| CN110361026B (en) | Humanoid robot path planning method based on 3D point cloud | |
| KR20050028859A (en) | Apparatus and method for recognizing environment, apparatus and method for designing path, and robot apparatus | |
| Kanoulas et al. | Footstep planning in rough terrain for bipedal robots using curved contact patches | |
| CN112975978B (en) | Multi-legged robot load balancing method and device and multi-legged robot | |
| CN112327899B (en) | Variable-configuration quadruped robot motion control method and system | |
| US10196104B1 (en) | Terrain Evaluation for robot locomotion | |
| CN104606886B (en) | A kind of method of distributed collisionless movement planning | |
| CN116619382A (en) | Robot motion control method and system and electronic equipment | |
| CN110928296A (en) | Method for avoiding charging seat by robot and robot thereof | |
| JP2012103144A (en) | Three-dimensional environment restoration device, three-dimensional environment restoration method, and robot | |
| Chestnutty et al. | Safe adjustment regions for legged locomotion paths | |
| JP5310236B2 (en) | Legged robot and its control method | |
| CN107914271A (en) | A kind of control method and control system based on robot | |
| CN110497414A (en) | The continuous torso exercise method for planning track of quadruped robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190205 |