CN105599821A - Electric driven biomimetic four-leg robot with environment sensing ability and control method - Google Patents

Electric driven biomimetic four-leg robot with environment sensing ability and control method Download PDF

Info

Publication number
CN105599821A
CN105599821A CN201610006061.8A CN201610006061A CN105599821A CN 105599821 A CN105599821 A CN 105599821A CN 201610006061 A CN201610006061 A CN 201610006061A CN 105599821 A CN105599821 A CN 105599821A
Authority
CN
China
Prior art keywords
trunk
sensor
environment sensing
robot
electric drive
Prior art date
Application number
CN201610006061.8A
Other languages
Chinese (zh)
Other versions
CN105599821B (en
Inventor
范永
谢爱珍
Original Assignee
山东优宝特智能机器人有限公司
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 山东优宝特智能机器人有限公司 filed Critical 山东优宝特智能机器人有限公司
Priority to CN201610006061.8A priority Critical patent/CN105599821B/en
Publication of CN105599821A publication Critical patent/CN105599821A/en
Application granted granted Critical
Publication of CN105599821B publication Critical patent/CN105599821B/en

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B62LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62DMOTOR VEHICLES; TRAILERS
    • B62D57/00Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track
    • B62D57/02Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members
    • B62D57/032Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/04Programme control other than numerical control, i.e. in sequence controllers or logic controllers

Abstract

The invention discloses an electric driven biomimetic four-leg robot with the environment sensing ability and a control method. A head is hinged to a front body through a neck and a head connecting rod, the head is linked with the neck, and the front body is hinged to a back body; a tail is arranged at the back part of the back body, and is also hinged to the front body through a tail connecting rod; front legs are respectively arranged on the two sides of the front body; back legs are respectively arranged on the two sides of the back body; the length of the back legs are larger than that of the front legs; the widest part of the front body is wider than the widest part of the back body; the electric driven biomimetic four-leg robot further comprises an environment sensing sensor group and a camera; the environment sensing sensor group and the camera are connected with a central processing unit; the output end of the central processing unit is connected with an electric driven execution system; a power management module supplies power for the environment sensing sensor group, the central processing unit and the electric driven execution system. Through the adoption of the electric driven control mode, the biomimetic four-leg robot can realize dynamic sensing, autonomous navigation and real-time following on environment, and can adapt to biomimetic gait motion planning with topographical changes and random disturbance.

Description

There is the bionical quadruped robot of electric drive and the control method of environment sensing ability

Technical field

The present invention relates to bio-robot technical field, relate in particular to a kind of electric drive with environment sensing ability bionicalQuadruped robot and control method.

Background technology

From robot motion mode, robot is broadly divided into wheeled robot, caterpillar type robot and leg foot formulaRobot three classes. With respect to wheeled or caterpillar type robot, the motion of leg foot formula has wheeled or crawler type is not available superiorCharacteristic:

(1) foot point that falls of leg foot formula motion is discrete, can in reached at the space of foot, select the optimum strong point, canLeaping over obstacles, has improved the adaptive capacity to rugged ground.

(2) motion of leg foot formula can realize all-around mobile, moves more flexible.

(3) motion of leg foot formula can active vibration isolation, i.e. trunk center of mass motion track and the decoupling zero of foot end track, at ground heightHealth still can held stationary at ordinary times.

(4) motion of leg foot formula still can realize higher translational speed on unevenness or loose ground, and energy consumption can be not remarkableIncrease.

Current, in world wide, some countries have launched research widely to quadruped robot, have obtained plentiful and substantial one-tenthReally, after especially boston, u.s.a power (BostonDynamics) has been issued hydraulic pressure quadruped robot BigDog, in the worldIn scope, start the upsurge of research hydraulic pressure quadruped robot, as Industrial Technology Research Institute of Korea S has researched and developed fluid motor-drivenQuadruped robot, the Italian Institute of Technology has researched and developed the quadruped robot HyQ of electric liquid combination drive, and Boston power is sent out again subsequentlyCloth AlphaDog and LS3 quadruped robot, aspect disturbance rejection, load capacity and adaptive capacity to environment, having significantly and carryingRise.

At home, the colleges and universities such as Tsing-Hua University, Shandong University, Harbin Institute of Technology and Central China University of Science and Technology are also being engaged in fourThe R&D work of foot bio-robot, and obtained certain achievement.

From the current study, the type of drive of quadruped robot is divided into hydraulic-driven and the large class of electric drive two.

Hydraulic-driven quadruped robot is possessing its advantage aspect high dynamic, high load capability, but exist power consumption large andThe problems such as noise is large.

The free degree of electric drive quadruped robot is relatively less, motion underaction; Perception to environment is poor, is difficult toAdapt to complicated ground motion.

Summary of the invention

Object of the present invention is exactly in order to address the above problem, and provides a kind of electric drive with environment sensing ability to imitateRaw quadruped robot and control method, simulation quadruped mammal figure, design head, trunk, shank and afterbody, integrated pass moreSensor, comprises vision sensor, sonac, infrared sensor, audio sensor, pressure sensor, three axial rake sensingsDevice, touch sensor etc., adopt electric drive control mode, realizes the dynamic sensing of bionical quadruped robot to environment, certainly leadingNavigate, follow in real time, and adaptation to the ground changes and the bionical gait motion planning of random disturbances.

To achieve these goals, the present invention adopts following technical scheme:

The bionical quadruped robot of electric drive with environment sensing ability, comprises head, and described head is by neck, headConnecting rod and front trunk are hinged, head and neck interlock, and described front trunk and rear trunk are hinged; The rear portion of described rear trunk is provided with tailPortion, described afterbody is also hinged by afterbody connecting rod and described front trunk, realizes the interlock of afterbody and front trunk; Described front trunkBoth sides are respectively equipped with foreleg, and the both sides of described rear trunk are respectively equipped with back leg; The length of described back leg is longer than foreleg; Described front bodyDry the widest part is wider than the widest part of rear trunk; Also comprise environment sensing sensor group and camera, described environment sensing sensingDevice group, camera are connected with central processing unit, and the output of described central processing unit connects electric drive executive system, power managementModule is described environment sensing sensor group, central processing unit and the power supply of electric drive executive system.

Described environment sensing sensor group comprise vision sensor, sonac, infrared sensor, audio sensor,Pressure sensor, three axial rake sensor and touch sensors.

One end in the two ends of described afterbody is provided with upper and lower two pin joints, described rear trunk and described afterbody upper hingedPoint articulated, one end of the lower link point of described afterbody and described afterbody connecting rod is hinged, the other end of described afterbody connecting rod with described inFront trunk is hinged.

Have the control method of the bionical quadruped robot of electric drive of environment sensing ability, environment sensing sensor group gathersExternal environment condition and self posture information, central processing unit is processed by decision-making level's strategy the environment sensing sensor group receiving and is adoptedThe information of collection, makes the reaction of owner's identification and foot end path planning, produces corresponding control signal control electric drive and carries out systemSystem performs an action.

Foot end paths planning method comprises, sets up four key point coordinate systems, by coordinate transformation equation, by robot weightHeart trajectory planning is transformed into foot end trajectory planning, selects different three dimensions tracks to obtain different movement effects, through anti-After conversion, obtain joint of robot exercise data, output control signal is to electric drive executive system.

Decision-making level's strategy processing need to be merged multiple sensors information, adopts the Weighted Average Algorithm that adds priority factorsAsk the result of decision; Sensor information after filtering, after normalization, first with priority factors computing, then obtain through weighted averageTo the last result of decision.

Owner knows method for distinguishing and comprises: owner's view data is made template, is stored in database, and camera collection is to mainAfter the image of people face, first pass through Kalman filtering, alleviate illumination, rotation, the impact of expression on face, then utilize integrationThe method of projection is determined face feature point, extracts the template of local feature, then carries out localized mode with owner's template of databasePlate coupling, judges whether it is owner.

Also comprise gesture control method, comprise two infrared sensors that left and right arranges, the each bifurcated of system mode is 4Kind, adopt quaternary tree ergodic algorithm, and insert time parameter in quaternary tree traversal; From a tree node toward next branches and leavesIn the process of node traversal, judge that time parameter is whether in limited range, walk continuing past next node, be less than scope and beBecause left and right sensor differences enters by mistake, need to jump to top-most node, go beyond the scope and get back to last layer node; Minimum oneNode layer represents the final result of gesture.

Also comprise the control method of tracking, the method that adopts infrared sensor and sonac to merge, with infrared sensingDevice testing result is as effective following range, and the Processing Algorithm of infrared sensor adopts quaternary tree ergodic algorithm binding time to becomeAmount adds sonac Processing Algorithm integer programming algorithm in effective following range, and hyperacoustic range data is as changeAmount, decision factor is a fixed value of setting through debugging, the result of product of range data and decision factor is as motion knotFruit reference quantity, motion result reference quantity finally obtains motion result through cutting plane algorithm.

Walking adopts sliding steps gait planning, and balance control strategy is selected the stability margin in robot stabilization criterion, whenWhen stability margin exceeds stability range value, in conjunction with squatting down tactful and waving arm strategy and adjust the attitude of robot, until recover steadyDetermine state.

Beneficial effect of the present invention:

The present invention is based on the identification of infrared sensing obstacle, realize " gesture control " function, can either retrieve accurately gestureAs a result, make the time range of gesture motion wider, also can solve the detection of left and right sensor apart from inconsistence problems.

The present invention contains multiple sensors, adopts electric drive control mode, realizes bionical quadruped robot moving environmentState perception, independent navigation, follow in real time, and adaptation to the ground changes and the bionical gait motion of random disturbances is planned.

Brief description of the drawings

Fig. 1 is top view of the present invention;

Fig. 2 is front view of the present invention;

Fig. 3 is the foreleg schematic diagram of standing;

Fig. 4 is the back leg schematic diagram of standing;

Fig. 5 is axonometric drawing of the present invention;

Fig. 6 is the structural representation of power management module;

Fig. 7 is control system structure chart of the present invention;

Fig. 8 is environment sensing system diagram;

Fig. 9 is electric drive executive system.

Wherein, 1. head, 2. front trunk, 3. rear trunk, 4. afterbody, 5. head connecting rod, 6. neck, 7. afterbody connecting rod, 8.Waist, 9. foreleg, 10. back leg, 11. elastic damping bodies, 12. first forearms, 13 second forearms, 14 first postbrachiums, after 15 secondArm,

9.1 shoulder yaw joints, 9.2 shoulder pitching joints, 9.3 ancon pitching joints, 9.4 wrist joints;

10.1 hip yaw joints, 10.2 hip knee pitching joints, 10.3 ankle-joints, 10.4 articulations digitorum manus.

Detailed description of the invention

Below in conjunction with accompanying drawing and embodiment, the invention will be further described.

As shown in Figure 1-2, there is the bionical quadruped robot of electric drive of environment sensing ability, comprise head 1, described head2 is hinged by neck 6, head connecting rod 5 and front trunk 2, and head 1 and neck 6 link, and described front trunk 2 is hinged with rear trunk 3;The rear portion of described rear trunk 3 is provided with afterbody 4, and described afterbody 4 is also hinged by afterbody connecting rod 7 and described front trunk 2, realizes afterbody4 with the interlock of front trunk 2; The both sides of described front trunk 2 are respectively equipped with foreleg 9, and the both sides of described rear trunk 3 are respectively equipped with back leg10; The length of described back leg 10 is longer than foreleg 9; The widest part of described front trunk 2 is wider than the widest part of rear trunk 3; Also comprise ringBorder detecting sensor group and camera, described environment sensing sensor group, camera are connected with central processing unit, described centreThe output of reason device connects electric drive executive system, and power management module is described environment sensing sensor group, central processing unitAnd electric drive executive system power supply.

Described environment sensing sensor group comprise vision sensor, sonac, infrared sensor, audio sensor,Pressure sensor, three axial rake sensor and touch sensors.

The present embodiment provides a kind of concrete structure of robot.

As shown in Figure 1-2, front trunk 2 comprises the front trunk front portion joining successively, front trunk middle part and front trunk rear portion, instituteThe width of stating front trunk front portion is less than the width at front trunk rear portion, and the width at described front trunk rear portion is less than the wide of front trunk middle partDegree. Described rear trunk comprises the rear trunk front portion joining successively, rear trunk middle part and rear trunk rear portion, described rear trunk rear portionWidth is less than the width of rear trunk front portion, and the width of described rear trunk front portion is less than the width at rear trunk middle part.

Described front trunk rear portion and rear trunk front portion are hinged, form waist 8. Entirety is that shoulder breadth hip is narrow like this, in addition waist 8The free degree, totally present the thin structure in front wide and rear narrow centre, make figure more graceful. Front and back trunk designs separately, passes throughWaist 8 frees degree connect, the relative elevating movement of trunk before and after waist 8 frees degree can realize, make mass motion more coordinate,Stable.

Described front trunk is anterior all hinged with one end of described neck 6 and one end of head connecting rod 5, described neck 6 anotherThe other end of one end and head connecting rod 5 is all hinged with head 1. The two ends of neck 6 and head connecting rod 5 are separately fixed at head 1 HeFront trunk, its four pin joints form the sealing quadrangle of four frees degree, free with the active of front trunk junction by neck 6Degree is realized neck interlock effect. The advantage of head connecting rod 5 mechanisms is: can ensure in the time regulating head 1 sensor relative position,Do not change the attitude of the relative trunk of sensor, ensure the effective range of measurement field with respect to trunk.

One end in the two ends of described afterbody 4 is provided with upper and lower two pin joints, described rear trunk rear portion and described afterbody 4Upper link point is hinged, and one end of the lower link point of described afterbody 4 and described afterbody connecting rod 7 is hinged, another of described afterbody connecting rod 7End is hinged with described front trunk rear portion. Afterbody 4 is by afterbody connecting rod 7 and waist 8 free degree interlocks. The two-end-point of afterbody connecting rod 7Be hinged on respectively front trunk and afterbody 4, afterbody 4 and the pin joint of rear trunk 3 and the pin joint of waist 8 in addition, forms four certainlyBy the sealing quadrangle of spending, realized the interlock effect of afterbody 4 by the waist 8 active frees degree. Can be according to the moving forward and backward of head 1,And drive relatively moving of afterbody 4 by waist 8, thereby keep overall barycenter stable in build motion and adjustment process.

As shown in Figure 3 and Figure 5, described foreleg 9 comprises shoulder yaw joint 9.1, the upper end in described shoulder yaw joint 9.1Hinged front trunk 2, lower end is by hinged the first forearm 12 in shoulder pitching joint 9.2, and described the first forearm 12 is by ancon pitchingHinged the second forearm 13 in joint 9.3, described the second forearm 13 is by the hinged front toe of wrist joint 9.4, described the second forearm 13 withBetween front toe, connect elastic damping body 11.

As shown in Figure 4 and Figure 5, described back leg 10 comprises hip yaw joint 10.1, described hip yaw joint 10.1The hinged rear trunk 3 in upper end, lower end is by hinged the first postbrachium 14 in hip knee pitching joint 10.2, and described the first postbrachium 14 closes by ankleSave hinged 10.3 second postbrachiums 15, described the second postbrachium 15 is by the hinged 10.4 rear toes of articulations digitorum manus, described the second postbrachium 15 withBetween rear toe, connect elastic damping body 11.

Copy quadruped mammal hip joint and shank skeletal structure, adopt foreleg 9 and the different scheme of back leg 10 sizes,By foreleg 9, the different connector of back leg 10, back leg 10 connectors are slightly long and become must angle, its back leg 10 driving powersLarger. Foreleg 9,10 of back legs all comprise three the initiatively free degree and driven frees degree; Aspect the topology of joint, adoptFull elbow formula joint topology, back leg 10 leg portion drive the structure that adopts hip knee one, pendulum scope after increasing, the ankle of back leg 10 closesJoint can play and regulate leg length and the auxiliary effect of having an effect. Its advantage is the working space that has increased and optimized shank,Foot end possesses larger and effective range of movement, is more applicable to walking or running at a high speed, adopts more bionical joint topologyStructure, the figure that makes to stand is more attractive in appearance, lively; And when motion, foreleg 9 turns to and assistive drive, and back leg 10 drives, and has improved fortuneMoving mobility, and make action more freely, bionical.

Foreleg 9 of the present invention, back leg 10 adopt the elastic shock attenuation body of exclusive articulations digitorum manus, and the wrist that is arranged on respectively foreleg 2 closesBetween joint and front toe, and between the articulations digitorum manus of back leg 10 and rear toe. By elastic shock attenuation body and driven wrist joint fromBy degree and the articulations digitorum manus free degree, reduce the shock on shank and ground, play protection driver, keep the work of trunk motion stabilizationWith.

The lower end in described back leg 10 hip knee pitching joints tilts to afterbody 4 directions as shown in Figure 3. For back leg 10 is than foreleg 2The part growing provides space.

Described elastic damping body 11 is Compress Spring, rubber spring, complex spring, oil gas or air spring.

The head cervical of robot is made up of two frees degree: one can left rotation and right rotation a joint, one can be in the pastThe neck joint of rear rotation.

The backbone of robot comprises one degree of freedom: can realize and bend over, swing back.

Each shank of robot is made up of three degree of freedom: can side direction rotation hip joint, can before and after rotationHip joint and knee joint, robot foot end has three-dimensional reached at a space like this, is reducing the complexity of quadruped robotWhen property, the shank of robot can complete the action such as swing, sideshake, and Zhe Shi robot can complete smoothlyThe action such as walk, turn to, and can in rugged landform, walk.

The afterbody of robot has one degree of freedom: realize swinging up and down of tail.

Robot head, trunk, leg foot, afterbody totally 16 frees degree, driven by motor respectively. Electric drive executive system bagDraw together motor, decelerator, four parts of position probing.

Motor is main execution unit, and range of movement, from 0 ° to 270 °, has ensured space that can be movable, meets robotFree degree requirement. Structural design combines motor and shank, is designed to integrated shank parts, is easy to install change.The motor of selection of Motor band position sensor, does not need to increase extra sensor.

Adopt PWM mode to drive. Electric drive executive system as shown in Figure 9.

As shown in Figure 6, described power management module comprises that voltage detection module, current detection module and electric weight detect mouldPiece, electric weight detection module is connected with battery, current detection module, voltage detection module; Voltage detection module is by the 3rd isolationModule is connected with described central processing unit; Electric weight detection module connects the first regulating filtering module, and the first regulating filtering module is logicalCross the first isolation module and be connected with described central processing unit, electric weight detection module is held by the second regulating filtering module and electric driveRow system connects, and electric drive executive system is connected with described central processing unit by the second isolation module.

Robot adopts 7.2V lithium battery power supply. Capacity is 3500mAh, and only 68*38*18mm of battery volume, is ensureingProvide enough electric weight whiles to system, saved and taken up room, make robot can increase in inside more functional module.

Robot is under some specific gait, and operating current can increase suddenly, and with high-frequency harmonic, therefore system is establishedCount power-supply management system, comprised the circuit such as voltage detecting, current detecting, electric weight detection, voltage stabilizing design, harmonic wave absorption. Power supplyManagement system has also comprised isolation design, by the isolated from power of control system and kinematic system, guarantees the power supply ripple of kinematic systemThe moving control system that can not affect. Power management part block diagram is as shown in 6.

Voltage detecting major function is monitoring system voltage, and alarm signal is provided in the time of electric voltage exception. Due to what monitor beBattery voltage signal, therefore the linear opto-coupler chip of the 3rd isolation module choice set is isolated, and uses little peripheral circuitRealize the isolation detection of voltage.

Current detection circuit monitoring system electric current for example, provides guard signal in the time running into abnormal (motor rotation blockage) situation,Parting system power supply, other device of protection system. Use resettable fuse cheaply, in the time of system overcurrent, automatically cut offPower supply circuits, play a protective role; System power recovers normal can continuation and uses later, does not need frequently to change fuse.

Electric quantity detection function provides charge information in time to user when being protection battery, and information about power offers automaticallyCharging system, ensures robot normal operation continuously in the situation that of not power-off. Electric weight detects the integrated chip that uses TI company,Play the effect of simplifying circuit design.

Voltage stabilizing designs the voltage in safe range to drive motors is provided, and protection motor is (firm under the different conditions of batteryBe full of electricity or use a period of time) can safe operation, can not be subject to the impact of cell voltage. Harmonic wave absorbs design and absorbs fortuneThe high-frequency harmonic that moving system produces, protection battery and interlock circuit.

The second isolation module, has adopted multi-channel high-speed photoelectric coupled device, the little encapsulation of selecting Japanese NEC Corporation to produceChip PS2805, meets the demand that circuit integration designs.

Have the control method of the bionical quadruped robot of electric drive of environment sensing ability, environment sensing sensor group gathersExternal environment condition and self posture information, central processing unit is processed by decision-making level's strategy the environment sensing sensor group receiving and is adoptedThe information of collection, makes the reaction of owner's identification and foot end path planning, produces corresponding control signal control electric drive and carries out systemSystem performs an action.

Robot is by multi-sensor collection external environment condition and self posture information, and controller is by locating heat transfer agentReason is carried out path and motion planning, produces corresponding control signal and outputs to electric drive performance element, controls the corresponding free degreeMotion. Control system entire block diagram as shown in Figure 7.

Control system central processing unit is selected the double-core CPU of DSP+ARM, and DSP kernel is responsible for algorithm computing, for example Zu Duan roadFootpath planning algorithm, sensing system related algorithm etc.; ARM kernel is responsible for collection, conversion, communication, the control letter of sensor informationThe work such as number output, the processing of decision-making level strategy. Double-core CPU takes full advantage of the operational capability of DSP and the disposal ability of ARM, bothCan make up the deficiency of monokaryon CPU aspect different disposal ability, also can reduce control system complexity, reduce control systemVolume.

Foot end paths planning method, sets up four key point coordinate systems, in the present embodiment, taking robot center of gravity as coordinate formerPoint is set up (X0, Y0, Z0) coordinate system, sets up (X1, Y1, Z1) coordinate system taking trunk hip joint as the origin of coordinates, closes with thigh hipJoint is set up (X2, Y2, Z2) coordinate system for the origin of coordinates, sets up (X3, Y3, Z3) coordinate system taking knee joint as the origin of coordinates, passes throughCoordinate transformation equation, is transformed into foot end trajectory planning by robot barycenter trajectory planning, selects different three dimensions tracks to obtainTo different movement effects, after inverse transformation, obtain joint of robot exercise data, output control signal is held to electric driveRow system.

Decision-making level's strategy processing need to be merged multiple sensors information, adopts the Weighted Average Algorithm that adds priority factorsAsk the result of decision; Sensor information after filtering, after normalization, first with priority factors computing, then obtain through weighted averageTo the last result of decision.

Multiple sensors that robot is integrated, comprises vision sensor, sonac, infrared sensor, audio frequency sensingDevice, pressure sensor, three axial rake sensors, as shown in Figure 8. By the fusion to these sensor informations, robot is realThe now detection to external environment condition, to the perception of self pose, movable information, and can make the anti-of simulating human according to environmentShould, for example identify owner, gesture control, follow the tracks of, keep away barrier, the behavior such as tracking, the walking of rugged road surface, voice answer-back, realize machineIn people definition about the part of " people ".

Owner knows method for distinguishing and comprises: owner's view data is made template, is stored in database, and camera collection is to mainAfter the image of people face, first pass through Kalman filtering, alleviate illumination, rotation, the impact of expression on face, then utilize integrationThe method of projection is determined face feature point, extracts the template of local feature, then carries out localized mode with owner's template of databasePlate coupling, judges whether it is owner.

Also comprise gesture control method, comprise two infrared sensors that left and right arranges, the each bifurcated of system mode is 4Kind, adopt quaternary tree ergodic algorithm, and insert time parameter in quaternary tree traversal; From a tree node toward next branches and leavesIn the process of node traversal, judge that time parameter is whether in limited range, walk continuing past next node, be less than scope and beBecause left and right sensor differences enters by mistake, need to jump to top-most node, go beyond the scope and get back to last layer node; Minimum oneNode layer represents the final result of gesture. The method can either retrieve gesture result accurately, makes the time model of gesture motionEnclose widelyr, also can solve the detection of left and right sensor apart from inconsistence problems.

Also comprise the control method of tracking, the method that adopts infrared sensor and sonac to merge, with infrared sensingDevice testing result is as effective following range, and the Processing Algorithm of infrared sensor adopts quaternary tree ergodic algorithm binding time to becomeAmount adds sonac Processing Algorithm integer programming algorithm in effective following range, and hyperacoustic range data is as changeAmount, decision factor is a fixed value of setting through debugging, the result of product of range data and decision factor is as motion knotFruit reference quantity, motion result only has five kinds of integer values: advance, retreat, turn left, right-hand rotation, as you were, motion result reference quantity warpCross cutting plane algorithm and finally obtain motion result.

Based on the processing of voice sensing, realize " speech recognition is replied ", select to have the core of unspecified person speech recognition technologySheet LD3320, master control ARM and identification chip adopt spi bus to communicate, and master control ARM writes control command toward identification chip,For example use voice command, the bootrom of Chinese phonetic alphabet setting to start to identify etc. Identification sensor is connected to identification coreAbove sheet, the process of identification is completed by chip internal algorithm, and result is sent to master control ARM. Master control ARM completes recognition resultResponse is processed, and comprises action voice answer-back (unspecified person sends " advancing " voice command, and robot starts forward motion), switchesSystem control model (for example by gesture control model convert to keep away barrier walking mode), voice output reply (robot play rightAnswer voice, music) etc.

Based on pressure sensing and inclination angle sensing processing, realize " submissive control " and " statokinetic ". In conjunction with walking and machinePeople's balance control policy control. What walking adopted is sliding steps gait planning, and balance control strategy selects robot stabilization to sentenceAccording in stability margin, three axle attitude angle determine stability margin as state variable. When stability margin exceeds stability range valueTime, in conjunction with squatting down tactful and waving arm strategy and adjust the attitude of robot, until recover stable state.

By reference to the accompanying drawings the specific embodiment of the present invention is described although above-mentioned, not the present invention is protected to modelThe restriction of enclosing, one of ordinary skill in the art should be understood that, on the basis of technical scheme of the present invention, those skilled in the art are notNeed to pay various amendments that creative work can make or distortion still in protection scope of the present invention.

Claims (10)

1. the bionical quadruped robot of electric drive with environment sensing ability, comprises head, it is characterized in that, described head passes through neckPortion, head connecting rod and front trunk are hinged, head and neck interlock, and described front trunk and rear trunk are hinged; After described rear trunkPortion is provided with afterbody, and described afterbody is also hinged by afterbody connecting rod and described front trunk, realizes the interlock of afterbody and front trunk; DescribedThe both sides of front trunk are respectively equipped with foreleg, and the both sides of described rear trunk are respectively equipped with back leg; The length of described back leg is longer than foreleg;The widest part of described front trunk is wider than the widest part of rear trunk; Also comprise environment sensing sensor group and camera, described environmentDetecting sensor group, camera are connected with central processing unit, and the output of described central processing unit connects electric drive executive system,Power management module is described environment sensing sensor group, central processing unit and the power supply of electric drive executive system.
2. the bionical quadruped robot of electric drive as claimed in claim 1 with environment sensing ability, is characterized in that described environmentDetecting sensor group comprises that vision sensor, sonac, infrared sensor, audio sensor, pressure sensor, three axles inclineAngle transducer and touch sensor.
3. electric drive four-leg bionic robot as claimed in claim 1, is characterized in that, the one end in the two ends of described afterbody is provided withUpper and lower two pin joints, the upper link point of described rear trunk and described afterbody is hinged, the lower link point of described afterbody and described tailOne end of portion's connecting rod is hinged, and the other end of described afterbody connecting rod and described front trunk are hinged.
4. the control method of the bionical quadruped robot of electric drive based on thering is environment sensing ability described in claim 2, its spyLevy and be, environment sensing sensor group gathers external environment condition and self posture information, and central processing unit is processed by decision-making level's strategyThe information that the environment sensing sensor group receiving gathers, makes the reaction of owner's identification and foot end path planning, produces correspondingControl signal control electric drive executive system perform an action.
5. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureThat foot end paths planning method comprises, sets up four key point coordinate systems, by coordinate transformation equation, by robot center of gravity railMark planning is transformed into foot end trajectory planning, selects different three dimensions tracks to obtain different movement effects, through inverse transformationObtain joint of robot exercise data, output control signal is to electric drive executive system later.
6. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureBe, decision-making level's strategy processing need to be merged multiple sensors information, adopts the Weighted Average Algorithm that adds priority factors to ask certainlyPlan result; Sensor information after filtering, after normalization, first with priority factors computing, then obtain through weighted averageAfter the result of decision.
7. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureBe that owner knows method for distinguishing and comprises: owner's view data is made template, be stored in database, camera collection is to owner's faceAfter the image of portion, first pass through Kalman filtering, alleviate illumination, rotation, the impact of expression on face, then utilize integral projectionMethod determine face feature point, extract the template of local feature, then carry out local template with owner's template of databaseJoin, judge whether it is owner.
8. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureBe, also comprise gesture control method, comprise two infrared sensors that left and right arranges, the each bifurcated of system mode is 4 kinds, adoptsQuaternary tree ergodic algorithm, and insert time parameter in quaternary tree traversal; From a tree node toward next branches and leaves node timeIn the process of going through, judge that time parameter whether in limited range, is continuing to walk toward next node, being less than scope is due to a left sideRight sensor differences enters by mistake, need to jump to top-most node, goes beyond the scope and gets back to last layer node; A minimum node layerRepresent the final result of gesture.
9. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureBe, also comprise the control method of tracking, the method that adopts infrared sensor and sonac to merge, detects with infrared sensorResult is as effective following range, and the Processing Algorithm of infrared sensor adopts quaternary tree ergodic algorithm binding time variable, is havingEffect adds sonac Processing Algorithm integer programming algorithm in following range, and hyperacoustic range data is as variable, decision-makingThe factor is a fixed value of setting through debugging, and the result of product of range data and decision factor is as motion result referenceAmount, motion result reference quantity finally obtains motion result through cutting plane algorithm.
10. there is as claimed in claim 4 the control method of the bionical quadruped robot of electric drive of environment sensing ability, its featureBe, walking adopts sliding steps gait planning, and balance control strategy is selected the stability margin in robot stabilization criterion, abundant when stablizingWhen amount exceeds stability range value, in conjunction with squatting down tactful and waving arm strategy and adjust the attitude of robot, until recover stable state.
CN201610006061.8A 2016-01-06 2016-01-06 The bionical quadruped robot of electric drive and control method with environment sensing ability CN105599821B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610006061.8A CN105599821B (en) 2016-01-06 2016-01-06 The bionical quadruped robot of electric drive and control method with environment sensing ability

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610006061.8A CN105599821B (en) 2016-01-06 2016-01-06 The bionical quadruped robot of electric drive and control method with environment sensing ability

Publications (2)

Publication Number Publication Date
CN105599821A true CN105599821A (en) 2016-05-25
CN105599821B CN105599821B (en) 2019-03-01

Family

ID=55980301

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610006061.8A CN105599821B (en) 2016-01-06 2016-01-06 The bionical quadruped robot of electric drive and control method with environment sensing ability

Country Status (1)

Country Link
CN (1) CN105599821B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107045552A (en) * 2017-04-28 2017-08-15 齐鲁工业大学 A kind of quadruped robot motion control method and control device based on sinusoidal diagonal gait Yu quick look-up table
CN108415330A (en) * 2018-03-13 2018-08-17 哈尔滨理工大学 Bio-robot foot end design method
CN110262532A (en) * 2019-06-24 2019-09-20 重庆大学 A kind of robot landform processing and more landform gait control method and system

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1476589A (en) * 2001-08-23 2004-02-18 索尼公司 Robot apparatus, face recognition method and face recognition apparatus
CN101770235A (en) * 2009-01-01 2010-07-07 索尼株式会社 Path planning device, path planning method, and computer program
CN102637036A (en) * 2012-05-08 2012-08-15 北京理工大学 Combined type bionic quadruped robot controller
JP2014161991A (en) * 2013-02-28 2014-09-08 Nsk Ltd Robot movement mechanism and robot comprising the same
CN204110199U (en) * 2014-09-05 2015-01-21 西南科技大学 A kind of running gear and adopt the bio-robot of this device
US20150107915A1 (en) * 2013-10-21 2015-04-23 Areva Inc. Vacuum Stepper Robot
CN104554510A (en) * 2015-01-04 2015-04-29 武汉理工大学 Bionic robot dog with flexible structure
CN204546536U (en) * 2015-04-19 2015-08-12 张仁良 A kind of home-services robot

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1476589A (en) * 2001-08-23 2004-02-18 索尼公司 Robot apparatus, face recognition method and face recognition apparatus
CN101770235A (en) * 2009-01-01 2010-07-07 索尼株式会社 Path planning device, path planning method, and computer program
CN102637036A (en) * 2012-05-08 2012-08-15 北京理工大学 Combined type bionic quadruped robot controller
JP2014161991A (en) * 2013-02-28 2014-09-08 Nsk Ltd Robot movement mechanism and robot comprising the same
US20150107915A1 (en) * 2013-10-21 2015-04-23 Areva Inc. Vacuum Stepper Robot
CN204110199U (en) * 2014-09-05 2015-01-21 西南科技大学 A kind of running gear and adopt the bio-robot of this device
CN104554510A (en) * 2015-01-04 2015-04-29 武汉理工大学 Bionic robot dog with flexible structure
CN204546536U (en) * 2015-04-19 2015-08-12 张仁良 A kind of home-services robot

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107045552A (en) * 2017-04-28 2017-08-15 齐鲁工业大学 A kind of quadruped robot motion control method and control device based on sinusoidal diagonal gait Yu quick look-up table
CN107045552B (en) * 2017-04-28 2020-04-24 齐鲁工业大学 Motion control method and control device of quadruped robot based on sine diagonal gait and rapid table look-up method
CN108415330A (en) * 2018-03-13 2018-08-17 哈尔滨理工大学 Bio-robot foot end design method
CN110262532A (en) * 2019-06-24 2019-09-20 重庆大学 A kind of robot landform processing and more landform gait control method and system

Also Published As

Publication number Publication date
CN105599821B (en) 2019-03-01

Similar Documents

Publication Publication Date Title
Hutter et al. Toward combining speed, efficiency, versatility, and robustness in an autonomous quadruped
Englsberger et al. Overview of the torque-controlled humanoid robot TORO
JP6227343B2 (en) Robot walking control method
Gienger et al. Task-oriented whole body motion for humanoid robots
Miller et al. Motion capture from inertial sensing for untethered humanoid teleoperation
CN100436237C (en) Human-imitating double-foot robot artificial leg
US6493606B2 (en) Articulated robot and method of controlling the motion of the same
Kanoun et al. Planning foot placements for a humanoid robot: A problem of inverse kinematics
Silva et al. Energy analysis during biped walking
Chevallereau et al. Optimal reference trajectories for walking and running of a biped robot
CN106573377B (en) Humanoid robot with collision avoidance and trajectory recovery capabilities
Culha et al. Quadrupedal bounding with an actuated spinal joint
JP2006051564A (en) Motion controlling device and method for robot device
Berns et al. Mechanical construction and computer architecture of the four-legged walking machine BISAM
Roussel et al. Generation of energy optimal complete gait cycles for biped robots
Dip et al. Genetic algorithm-based optimal bipedal walking gait synthesis considering tradeoff between stability margin and speed
Behnke Online trajectory generation for omnidirectional biped walking
Li et al. A passivity based admittance control for stabilizing the compliant humanoid COMAN
Stephens et al. Modeling and control of periodic humanoid balance using the linear biped model
Tellez et al. Reem-B: An autonomous lightweight human-size humanoid robot
Sabourin et al. Robustness of the dynamic walk of a biped robot subjected to disturbing external forces by using CMAC neural networks
Saranli et al. Template based control of hexapedal running
Srinivasan et al. A low-dimensional sagittal-plane forward-dynamic model for asymmetric gait and its application to study the gait of transtibial prosthesis users
Saidouni et al. Generating globally optimised sagittal gait cycles of a biped robot
Kim et al. Realization of dynamic walking for the humanoid robot platform KHR-1

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant