WO2017181976A1 - 提高六足机器人行走稳定性的步态规划方法 - Google Patents

提高六足机器人行走稳定性的步态规划方法 Download PDF

Info

Publication number
WO2017181976A1
WO2017181976A1 PCT/CN2017/081302 CN2017081302W WO2017181976A1 WO 2017181976 A1 WO2017181976 A1 WO 2017181976A1 CN 2017081302 W CN2017081302 W CN 2017081302W WO 2017181976 A1 WO2017181976 A1 WO 2017181976A1
Authority
WO
WIPO (PCT)
Prior art keywords
foot
fuselage
coordinate system
coordinates
joint
Prior art date
Application number
PCT/CN2017/081302
Other languages
English (en)
French (fr)
Inventor
孙天齐
Original Assignee
孙天齐
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 孙天齐 filed Critical 孙天齐
Publication of WO2017181976A1 publication Critical patent/WO2017181976A1/zh

Links

Images

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
    • 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
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D57/00—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track
    • B62D57/02—Vehicles 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
    • B62D57/032—Vehicles 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

Definitions

  • the invention relates to a hexapod robot, in particular to a gait planning method for improving the walking stability of a hexapod robot.
  • the hexapod robot also called the spider robot, is a kind of multi-legged robot. It has the advantages of good stability, high carrying capacity and strong terrain adaptability. It is the best choice in complex working environment and has broad application prospects.
  • the common gait of the hexapod robot is mainly based on the triangular gait.
  • the six feet are divided into two groups that are alternately separated. Each group has three feet.
  • the two groups of joints alternately rotate and rotate to realize the advancement, steering and retreat of the robot. .
  • the triangle gait can realize movement, rotation, etc., not only the action is monotonous, the rhythm of the action is strong, and the smooth movement cannot be smoothly and smoothly; and since the forward movement of the triangular gait is actually performed by alternate rotation of two different directions It is impossible to keep the front direction of the robot body stable and consistent. For a hexapod robot with a camera function, it will cause severe shaking of the video picture, which seriously affects the video quality and application range.
  • the technical problem to be solved by the present invention is that the triangular gait of the existing hexapod robot is not only monotonous and hard, but also has a strong rhythm of motion, cannot smoothly and smoothly move, and cannot maintain the problem that the front direction of the robot body is stable and consistent.
  • the technical solution adopted by the present invention is to provide a gait planning method for improving the walking stability of a hexapod robot, including the following basic actions:
  • the robot head camera module rotates to the direction of the target end point, and then performs the basic motion of travel.
  • the four-footed landing is always ensured when traveling, and the two adjacent feet are raised, including alternately lifting the two adjacent feet and the two adjacent feet alternately lifting the two gaits, wherein
  • the two adjacent feet are alternately raised as follows: after the start of the journey, the raised foot falls to the ground, and at the same time, the four feet of the ground are lifted up, and the raised foot is not adjacent to the original raised foot.
  • the translation path of each foot end is calculated by using a transformation algorithm to establish a correspondence relationship between the foot coordinate system of each foot and the body coordinate system, and in the standard standing posture, the translation path is calculated through the center of the body to calculate each a translational path at the end of the foot;
  • the calculation method of the panning path of the fuselage center is: in the fuselage coordinate system, the center of the fuselage is mapped to a line perpendicular to the ground through the target end point, and the center coordinates of the body map are obtained through the coordinates of the target end point, the body center and the fuselage
  • the mapping center connection is the panning path of the fuselage center.
  • the acquisition of the target end point coordinates is that the camera module searches for the target end point coordinates in the environment coordinate system, and then the target end point is projected to the body coordinate system coordinate in the environment coordinate system coordinate by the transformation algorithm.
  • the transformation algorithm Medium, obtain the coordinates of the target end point in the fuselage coordinate system.
  • the correspondence between the foot coordinate system of each foot and the body coordinate system is established by the transformation algorithm, and specifically includes the following steps:
  • ⁇ is the angle between the fuselage center and the camera module direction connection and the fuselage center and the foot root at the fuselage connection point;
  • the robot includes three joints of the hip joint, the knee joint and the ankle joint, and the angles of the three joints in the standard standing posture are as follows:
  • Hip joint perpendicular to the tangential direction of the contact point of the circumference of the fuselage
  • Knee joint parallel to the horizontal direction of the fuselage
  • Ankle joint Make the two limbs 90 degrees, and the ends of the feet are perpendicular to the ground.
  • the hip joint can move along the circumference of the body to rotate the body.
  • the preset angle of each joint rotation is calculated as: in each foot coordinate system, the joint angles of the robots are mapped with the coordinates of the foot end points, and the coordinate paths are used to calculate the translation paths of the respective foot ends. a motion trajectory on a vertical plane, and obtain a rotation angle of each joint corresponding to a different coordinate point of the motion trajectory at the end of the motion, and configure a preset rotation angle of the joint rotation;
  • the end of the foot is calculated by taking the D-H matrix corresponding to the rotation angle of each joint at different coordinate points of the motion trajectory.
  • the specific formula is:
  • ⁇ i represents the angle of rotation of each joint about the z-axis of its own coordinate system during motion
  • d i the offset between the joints on the z-axis during motion
  • i takes values from 1 to 3.
  • each foot performs a cyclic action according to the same cycle and different phase differences, and each foot smoothly alternately shifts;
  • the movement cycle of each foot is roughly divided into three parts: lifting, falling, and squatting; among them, the squatting part time accounts for 2/3 of the period.
  • the three-part motion of each foot's motion cycle is achieved by uniformly setting a plurality of key coordinate points at the end of each motion cycle, and then dividing the entire motion cycle into uniform according to the number of key coordinate points. A small cycle, then moving the end of the foot from one key coordinate point to the next key coordinate point in a small cycle.
  • the invention relies on a transformation algorithm to ensure that the end positions of the feet are always kept straight during the movement, so that the position of the body is also kept straight forward, and the ground position of the non-adjacent four feet is guaranteed to be avoided under any circumstances, and the action is frustrated, blunt, and the body.
  • Tilt, gait is smooth and stable; at the same time, only the head is rotated when rotating, the body does not move, keeping the front of the robot straight forward, without angular distortion, which can quickly change the front direction or move in a specified direction without moving the limb.
  • FIG. 1 is a schematic diagram of a trajectory planning method for improving the walking stability of a hexapod robot according to the present invention
  • FIG. 2 and FIG. 3 are schematic diagrams showing a hexapod end state of a gait planning method for improving walking stability of a hexapod robot according to the present invention
  • FIG. 4 is a schematic view showing the spatial relationship of each adjacent link of the robot in the present invention.
  • the invention is based on a hexapod robot with a circular symmetrical structure and a six-legged uniform distribution on the circumference of the fuselage, and the robot head disposed above the fuselage can be rotated 360 degrees, and the camera module is provided on the front face of the head.
  • the present invention defines three coordinate systems, as follows:
  • the fuselage coordinate system taking the center of the robot body as the origin, the line passing through the origin, parallel to the plane of the fuselage, perpendicular to the line in the front surface of the head is the Y-axis, parallel to the plane of the fuselage, perpendicular to the Y-axis is X
  • the axis, the line perpendicular to the plane of the fuselage is the Z axis;
  • the joint of each foot and the fuselage is taken as the coordinate origin, and the tangential line of the body through the coordinate origin is the X axis, and the extension line of the coordinate origin and the center of the fuselage is the Y axis, and the coordinate origin is perpendicular to the plane of the fuselage.
  • Straight line Z axis referred to as the foot coordinate system, a total of 6;
  • the ground plane on which the end of the robot foot is stepped is taken as the X-Y plane, and the origin can be set at any position on the X-Y plane, passing through the origin and perpendicular to the ground plane as the Z-axis.
  • the gait planning method for improving the walking stability of a hexapod robot mainly comprises the following basic actions:
  • the robot head camera module rotates to the direction of the target end point, and then performs the basic motion of the travel, without the need to rotate the body, thereby ensuring that the front direction of the robot body is stable and consistent. For the robot with camera function, avoid the video.
  • the robot head can be rotated 360 degrees.
  • the four-legged landing is always ensured during the traveling, and the non-adjacent two-legged lifting includes alternately lifting the two adjacent feet and the two adjacent feet alternately lifting the two gaits, wherein, alternately lifting The two feet that are not adjacent are: after the start of the journey, the two legs raised together land, and the two feet that are not adjacent to each other are lifted together; the two feet that are not adjacent are alternately raised: after the start of the journey, the lifted One foot falls, and at the same time, the four feet of the ground are lifted up, and the raised foot is not adjacent to the original raised foot.
  • the translation path of each foot end is calculated by using a transformation algorithm to establish a correspondence relationship between the foot coordinate system of each foot and the body coordinate system, and calculating the translation path of each foot end through the translational path of the center of the body (
  • the translational path of the hexapod end of the robot is a straight path, and the end of each foot is on the straight line, lifted and dropped.
  • the translation path of each end of the foot is calculated with reference to the position of the end of each foot in the standard standing posture.
  • the relative position of the end of the foot and the center of the fuselage is fixed, so when the fuselage is in any position, the coordinates of the end of each foot can be calculated, and the foot can be calculated through the translational path of the center of the fuselage.
  • the translational path of the end is fixed, so when the fuselage is in any position, the coordinates of the end of each foot can be calculated, and the foot can be calculated through the translational path of the center of the fuselage.
  • the calculation method of the panning path of the fuselage center is: in the fuselage coordinate system, the center of the fuselage is mapped to a line perpendicular to the ground through the target end point, and the center coordinates of the fuselage are obtained through the coordinates of the target end point, the fuselage center and the fuselage
  • the mapping center connection is the panning path of the fuselage center.
  • the acquisition of the target end point coordinates is obtained by the camera module to obtain the target end point coordinates in the environment coordinate system, and then the target end point is projected in the coordinate system of the environment coordinate system to the fuselage coordinate system by the transformation algorithm, and the machine is acquired.
  • the coordinates of the target end point in the body coordinate system is obtained by the camera module to obtain the target end point coordinates in the environment coordinate system, and then the target end point is projected in the coordinate system of the environment coordinate system to the fuselage coordinate system by the transformation algorithm, and the machine is acquired.
  • the transformation algorithm establishes the correspondence between the foot coordinate system of each foot and the body coordinate system.
  • the specific method is divided into two steps:
  • the robot includes three joints of the hip joint, the knee joint, and the ankle joint, and the angles of the three joints in the standard standing posture are as follows:
  • Hip joint perpendicular to the tangential direction of the contact point of the circumference of the fuselage
  • Knee joint parallel to the horizontal direction of the fuselage
  • Ankle joint Make the two limbs 90 degrees, and the ends of the feet are perpendicular to the ground.
  • the hip joint can move along the circumference of the body to rotate the body.
  • the hip joint movement is controlled. Go to the appropriate direction.
  • the preset angle of each joint rotation is calculated by mapping each joint angle of the robot with the coordinates of the foot end point in each foot coordinate system, and calculating the translation path of each foot end by coordinate mapping.
  • the motion trajectory on the vertical plane and obtain the rotation angle of each joint corresponding to the different end points of the motion trajectory of the foot, and configure the preset angle of the joint rotation.
  • the generation of the motion trajectory of each foot end in the vertical plane of its translational path can be initially set by experience.
  • the empirical setting is that we manually try to adjust the joint coordinates of each motion in the step and try repeatedly.
  • the series can be smoothly advanced, then improved and optimized based on it, and then the optimal solution is generated by genetic algorithm or other machine learning and optimization algorithms, and the optimal solution of the motion trajectory equation is imported into the robot storage.
  • the phase difference and period of the end motion of each foot can be generated by the CPG oscillator.
  • the phase difference and the period range theory are unrestricted.
  • the lower cycle time limit ie, the fastest speed
  • the lower cycle time limit is determined by the theoretical rotational speed of the motor, in the gait taken by the present invention.
  • the movement interval between each point is 250ms, that is, 3 seconds to complete a cycle.
  • the posture of the fuselage at the beginning and end of this cycle is exactly the same, but the position is Pan.
  • the rotation angle of each joint corresponding to the different end points of the motion trajectory is calculated by using a DH matrix
  • the DH matrix is a spatial relationship describing the adjacent links of the robot by using a 4x4 homogeneous matrix.
  • the central axis of each joint is the Z axis of each coordinate system
  • the coordinate system of the hip joint is the foot coordinate system
  • ⁇ i represents the rotation angle of each joint around the z coordinate of its own coordinate system during movement
  • ⁇ i is two adjacent motions when moving The torsion angle between the z-axis of the joint
  • d i represents the distance between two adjacent perpendicular lines on the z-axis during motion (or the offset between the joints)
  • a i represents the vertical line
  • the length of the link (the length of the link Link i ) Since the robot has three joints per foot, the value of i ranges from 1 to 3.
  • the specific formula for calculating the DH matrix is:
  • each leg performs a cyclic action according to the same cycle and different phase differences, so that each foot smoothly alternately advances, and the motion cycle of each foot is roughly divided into three parts: lifting, falling, and posterior, wherein the posterior temporal portion is part of the time.
  • the three-part motion of each foot's motion cycle is achieved by uniformly setting a plurality of key coordinate points at the end of each motion cycle, and then dividing the entire motion cycle into uniform small cycles according to the number of key coordinate points, and then In a small cycle, the end of the foot is moved from a key coordinate point to the next key coordinate point, thereby achieving the alternate movement of each foot.
  • the position of the key coordinate point is firstly designed by experience, and then optimized by an algorithm.

Abstract

一种提高六足机器人行走稳定性的步态规划方法，包括：行进：抬起不相邻两足，开始沿平行于机身中心和终点连线的每足末端的平移路径行进，行进时始终保证四足着地，不相邻两足抬起，且每足的各关节按周期旋转到对应预设角度，使每足末端在其平移路径上抬起、落下；变向：机器人头部摄像模块旋转到正对目标终点方向，再执行行进基本动作。采用此方法避免机器人动作顿挫、生硬、机身倾斜，步态流畅稳定；同时旋转时仅旋转头部，机身不动，保持机器人正面始终直线向前，不出现角度扭曲，能够实现不必移动肢体即可快速改变正面朝向或向指定方向移动。

Description

提高六足机器人行走稳定性的步态规划方法 技术领域
本发明涉及六足机器人，具体涉及提高六足机器人行走稳定性的步态规划方法。
背景技术
六足机器人又叫蜘蛛机器人，是多足机器人的一种，具有稳定性好、承载能力高、地形适应性强等优势，是复杂作业环境下的最佳选择，具有广阔的应用前景。
目前，六足机器人的常见步态以三角步态为主，即将六足分为交替相隔的两组，每组三足，通过两组关节交替起落旋转来实现机器人的前进、转向、后退等行动。三角步态虽然能实现移动、旋转等动作，不仅动作单调生硬、动作的节奏感较强，无法流畅平滑移动；而且由于三角步态的前进动作实际上是通过两个不同方向的交替旋转进行的，无法保持机器人机身正面方向稳定一致，对于有摄像头功能的六足机器人来说，会造成视频画面的剧烈晃动，严重影响视频质量和应用范围。
发明内容
本发明所要解决的技术问题是现有六足机器人的三角步态不仅动作单调生硬、动作的节奏感较强，无法流畅平滑移动，而且无法保持机器人机身正面方向稳定一致的问题。
为了解决上述技术问题，本发明所采用的技术方案是提供一种提高六足机器人行走稳定性的步态规划方法，包括以下基本动作：
行进：抬起不相邻两足，开始沿平行于机身中心和终点连线的每足末端的平移路径行进，行进时始终保证四足着地，不相邻两足抬起，且每足的各 关节按周期旋转到对应预设角度，使每足末端在其平移路径上抬起、落下；
变向：机器人头部摄像模块旋转到正对目标终点方向，再执行行进基本动作。
在上述方案中，行进时始终保证四足着地，不相邻两足抬起包括交替抬起不相邻的两足和不相邻的两足交替抬起两种步态，其中，
交替抬起不相邻的两足为：行进开始后，抬起的两足一起落地，着地、不相邻的两足一起抬起；
不相邻的两足交替抬起为：行进开始后，抬起的一足落地，同时着地的四足中一足抬起，且该抬起足与原抬起足不相邻。
在上述方案中，每足末端的平移路径计算方式为：利用变换算法建立每足的足坐标系与机身坐标系的对应关系，在标准站立姿态下，通过机身中心平移路径，计算出每足末端的平移路径；其中，
机身中心平移路径的计算方法为：在机身坐标系中，将机身中心映射到经过目标终点垂直于地面的直线上，通过目标终点坐标得到机身映射中心坐标，机身中心与机身映射中心连线为机身中心平移路径。
在上述方案中，在机身坐标系中，目标终点坐标的获取是摄像模块搜索得到在环境坐标系下的目标终点坐标，再通过变换算法将目标终点在环境坐标系坐标投射到机身坐标系中，获取在机身坐标系中目标终点坐标。
在上述方案中，通过变换算法建立每足的足坐标系与机身坐标系的对应关系具体包括以下步骤：
第一步、通过对机身坐标系坐标(x,y)进行旋转，得到足坐标系坐标(x’,y’)为：
Figure PCTCN2017081302-appb-000001
其中，θ为机身中心和摄像模块方向连线与机身中心和足根部在机身连接点连线的夹角；
第二步、对坐标(x’,y’)在足坐标系的y轴上平移r，r为机身半径，在x轴上不变，则得到机身坐标系坐标(x,y,z)在足坐标系的坐标(x”,y”,z”)；其中，平移后x”＝x’，y”＝y’+r，由于z轴未发生变换，所以z”＝z。
在上述方案中，机器人包括髋关节、膝关节和踝关节三个关节，标准站立姿态时三个关节的角度如下：
髋关节：与机身圆周接触点的切线方向垂直；
膝关节：与机身水平方向平行；
踝关节：使两段肢体成90度，足末端与地面垂直。
在上述方案中，所述髋关节能沿着机身圆周运动，使机身转动。
在上述方案中，各关节旋转的预设角度的计算方法为：在每个足坐标系中，将机器人各个关节角度与足末端点坐标进行映射，通过坐标映射计算出各个足末端在其平移路径竖直平面上的运动轨迹，并获得足末端在该运动轨迹不同坐标点上对应的各关节旋转角度，配置关节旋转预设角度；
足末端在运动轨迹不同坐标点上对应的各关节旋转角度采取D-H矩阵进行计算，具体公式为：
足末端的变换矩阵T为：[T]＝[Z1][Z2]...[Zn]；
相邻两连杆间的变换矩阵Zi为：
Figure PCTCN2017081302-appb-000002
其中，θi表示运动时各关节绕自身坐标系z轴的旋转角；di表示运动时在z轴上各关节之间的偏移；i取值为1至3。
在上述方案中，每足按相同周期、不同相位差进行循环动作，各足平稳交替平移；
每足的运动周期大致分为三部分：抬起、下落、后蹬；其中，后蹬部分时间占到周期的2/3。
在上述方案中，每足的运动周期的三部分动作的实现是通过将每个运动周期足末端的运动轨迹均匀设置多个关键坐标点，再根据关键坐标点数量将整个运动周期分为均匀的小周期，然后在一个小周期内将足末端从一个关键坐标点移到下一个关键坐标点。
本发明依赖变换算法保证各足末端位置在移动中始终保持直线，以使得机身位置也保持直线向前，且保证在任何情况下保证不相邻四足着地，避免动作顿挫、生硬、机身倾斜，步态流畅稳定；同时旋转时仅旋转头部，机身不动，保持机器人正面始终直线向前，不出现角度扭曲，能够实现不必移动肢体即可快速改变正面朝向或向指定方向移动。
附图说明
图1为本发明提供的一种提高六足机器人行走稳定性的步态规划方法每足的平移路径示意图；
图2、图3为本发明提供的一种提高六足机器人行走稳定性的步态规划方法的六足末端状态示意图；
图4为本发明中机器人各相邻连杆的空间关系示意图。
具体实施方式
本发明是基于机身为圆周对称结构、六足均匀分布于机身圆周的六足机器人，且设在机身之上的机器人头部可360度旋转，并在头部正脸带有摄像模块。为了方便说明，本发明定义了三个坐标系，具体如下：
机身坐标系，以机器人机身中心为原点，经过原点、平行于机身平面、垂直于头部正脸表面中分线的直线为Y轴，平行机身平面、垂直于Y轴直线为X轴，垂直于机身平面的直线为Z轴；
足坐标系，以每足根部与机身连接处为坐标原点，经坐标原点的机身切线为X轴，坐标原点与机身中心连接延长线为Y轴、经坐标原点垂直于机身平面的直线Z轴，简称足坐标系，共6个；
环境坐标系，以机器人足末端所踏的地平面作为X-Y平面，原点可以设置在X-Y平面任意位置上，经过原点并垂直于地平面方向为Z轴。
下面结合说明书附图和具体实施例对本发明做出详细的说明。
本发明提供的一种提高六足机器人行走稳定性的步态规划方法，主要包括以下基本动作：
行进：抬起不相邻两足，如图1所示，开始沿平行于机身中心和终点连线的每足末端的平移路径行进，行进时始终保证四足着地，不相邻两足抬起且每足的各关节按周期旋转到对应预设角度，使每足末端在其平移路径上抬起、落下，以便消除顿挫感，动作流畅平滑，按照此步态方案，六足末端状态仅存在图2、图3所示两种情况，其中空心圆表示抬起足的末端，黑色填充圆表示着地足的末端；
变向：机器人头部摄像模块旋转到正对目标终点方向，再执行行进基本动作，不需要机身进行旋转，从而保正机器人机身正面方向稳定一致，对于有摄像头功能的机器人来说，避免视频画面的剧烈晃动，在本发明中，机器人头部可进行360度旋转。
在本发明中，行进时始终保证四足着地，不相邻两足抬起包括交替抬起不相邻的两足和不相邻的两足交替抬起两种步态，其中，交替抬起不相邻的两足为：行进开始后，抬起的两足一起落地，着地、不相邻的两足一起抬起；不相邻的两足交替抬起为：行进开始后，抬起的一足落地，同时着地的四足中一足抬起，且该抬起足与原抬起足不相邻。
在本发明中，每足末端的平移路径计算方式为：利用变换算法建立每足的足坐标系与机身坐标系的对应关系，通过机身中心平移路径，计算出每足末端的平移路径(当机器人由一个坐标移动到另一个坐标点时，从俯视角度 来看，机器人六足末端平移路径为直线路径，每足末端在该直线上，抬起、落下)，这里以在标准站立姿态下各足末端的位置为参照，计算每足末端的平移路径，在标准站立姿态下，足末端和机身中心的相对位置是固定的，所以在机身位于任意位置时，都能算出各足末端的坐标，也就能通过机身中心平移路径计算出各足末端的平移路径。
机身中心平移路径的计算方法为：在机身坐标系中，将机身中心映射到经过目标终点垂直于地面的直线上，通过目标终点坐标得到机身映射中心坐标，机身中心与机身映射中心连线即为机身中心平移路径。
在机身坐标系中，目标终点坐标的获取是摄像模块搜索得到在环境坐标系下的目标终点坐标，再通过变换算法将目标终点在环境坐标系坐标投射到机身坐标系中，获取在机身坐标系中目标终点坐标。
通过变换算法建立每足的足坐标系与机身坐标系的对应关系具体方式分为两步：
旋转：设机身中心和摄像模块方向连线与机身中心和足根部在机身连接点连线的夹角为θ，则机身坐标系坐标(x,y)在足坐标系中的坐标(x’,y’)为：
Figure PCTCN2017081302-appb-000003
平移：旋转之后，令x”＝x’，y”＝y’+r，其中，r为机身半径，即机身中心到足根部的距离；由于z轴未发生变换，所以z”＝z，这样就将机身坐标系坐标(x,y,z)变换成了足坐标系(x”,y”,z”)，这样在足坐标系中通过机身中心点移动时各个坐标，就能推算出移动过程中各足末端的位置的对应坐标，每个足末端所有位置的对应坐标的连线即为每足末端的平移路径，如图1所示。
在本发明中，机器人包括髋关节、膝关节和踝关节三个关节，标准站立姿态时三个关节的角度如下：
髋关节：与机身圆周接触点的切线方向垂直；
膝关节：与机身水平方向平行；
踝关节：使两段肢体成90度，足末端与地面垂直。
在本发明中，髋关节能沿着机身圆周运动，使机身转动，当机器人机身上的固定机械手不能触碰到目标或其他需要旋转机身的情况时，控制髋关节运动，将机身转到相应方向。
在本发明中，各关节旋转的预设角度的计算方法为：在每个足坐标系中，将机器人各个关节角度与足末端点坐标进行映射，通过坐标映射计算出各个足末端在其平移路径竖直平面上的运动轨迹，并获得足末端在该运动轨迹不同坐标点上对应的各关节旋转角度，配置关节旋转预设角度。
在本发明中，各个足末端在其平移路径竖直平面上的运动轨迹的生成工作最初可依靠经验进行设置，经验设置是我们通过手动调节步伐中每一个动作的关节坐标，反复尝试，得出的系列可以平稳前进的动作，然后再以它为基础进行改进和优化，而后通过遗传算法或其他机器学习及优化算法生成最优解，并将运动轨迹方程最优解导入机器人存储。各足末端动作的相位差及周期可由CPG振荡器发生，相位差及周期范围理论是无限制的，周期时间下限(即最快速度)由电机的理论转速决定，在本发明采取的步态中，每个步伐周期中有12个参考坐标点，各点之间的移动间隔时间是250ms，也就是3秒完成一个周期循环，此周期起始和终结时的机身姿态完全相同，只是位置进行了平移。
在本发明中，足末端在运动轨迹不同坐标点上(足末端位置)对应的各关节旋转角度采取D-H矩阵进行计算，D-H矩阵是一种利用4x4齐次矩阵描述机器人各相邻连杆空间关系(如图2所示)，进而推导出机器人各连杆末端对于足坐标系位置的算法，机器人的每一足都可以看作一套连杆系统，在图2中，分别针对各关节建立坐标系，各个关节中心轴为每个坐标系的Z轴，髋关节的坐标系为足坐标系，θi表示运动时各关节绕自身坐标系z轴的旋 转角；αi为运动时两个相邻关节的z轴之间的扭角；di表示运动时在z轴上两条相邻的公垂线之间的距离(或称各关节之间的偏移)；ai表示各公垂线的长度(连杆Linki的长度)由于本机器人每足有三个关节，所以i取值范围为1至3，采取D-H矩阵进行计算的具体公式为：
相邻两连杆间的变换矩阵Zi为：
Figure PCTCN2017081302-appb-000004
足末端的变换矩阵T为：[T]＝[Z1][Z2]...[Zn]。
在本发明中，每足按相同周期、不同相位差进行循环动作，使得各足平稳交替前进，每足的运动周期大致分为三部分：抬起、下落、后蹬，其中，后蹬部分时间要占到周期的2/3，保证机器人始终有三足在地面上，形成平稳的步态。每足的运动周期的三部分动作的实现是通过将每个运动周期足末端的运动轨迹均匀设置多个关键坐标点，再根据关键坐标点数量将整个运动周期分为均匀的小周期，然后在一个小周期内将足末端从一个关键坐标点移到下一个关键坐标点，从而实现各足的交替移动，关键坐标点位置首先通过经验设计，之后通过算法优化。
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (10)

  1. 一种提高六足机器人行走稳定性的步态规划方法，其特征在于，包括以下基本动作：
    行进：抬起不相邻两足，开始沿平行于机身中心和终点连线的每足末端的平移路径行进，行进时始终保证四足着地，不相邻两足抬起，且每足的各关节按周期旋转到对应预设角度，使每足末端在其平移路径上抬起、落下；
    变向：机器人头部摄像模块旋转到正对目标终点方向，再执行行进基本动作。
  2. 如权利要求1所述的方法，其特征在于，行进时始终保证四足着地，不相邻两足抬起包括交替抬起不相邻的两足和不相邻的两足交替抬起两种步态，其中，
    交替抬起不相邻的两足为：行进开始后，抬起的两足一起落地，着地、不相邻的两足一起抬起；
    不相邻的两足交替抬起为：行进开始后，抬起的一足落地，同时着地的四足中一足抬起，且该抬起足与原抬起足不相邻。
  3. 如权利要求1所述的方法，其特征在于，每足末端的平移路径计算方式为：利用变换算法建立每足的足坐标系与机身坐标系的对应关系，在标准站立姿态下，通过机身中心平移路径，计算出每足末端的平移路径；其中，
    机身中心平移路径的计算方法为：在机身坐标系中，将机身中心映射到经过目标终点垂直于地面的直线上，通过目标终点坐标得到机身映射中心坐标，机身中心与机身映射中心连线为机身中心平移路径。
  4. 如权利要求3所述的方法，其特征在于，在机身坐标系中，目标终点坐标的获取是摄像模块搜索得到在环境坐标系下的目标终点坐标，再通过变换算法将目标终点在环境坐标系坐标投射到机身坐标系中，获取在机身坐标系中目标终点坐标。
  5. 如权利要求4所述的方法，其特征在于，通过变换算法建立每足的 足坐标系与机身坐标系的对应关系具体包括以下步骤：
    第一步、通过对机身坐标系坐标(x,y)进行旋转，得到足坐标系坐标(x’,y’)为：
    Figure PCTCN2017081302-appb-100001
    其中，θ为机身中心和摄像模块方向连线与机身中心和足根部在机身连接点连线的夹角；
    第二步、对坐标(x’,y’)在足坐标系的y轴上平移r，r为机身半径，在x轴上不变，则得到机身坐标系坐标(x,y,z)在足坐标系的坐标(x”,y”,z”)；其中，平移后x”＝x’，y”＝y’+r，由于z轴未发生变换，所以z”＝z。
  6. 如权利要求5所述的方法，其特征在于，机器人包括髋关节、膝关节和踝关节三个关节，标准站立姿态时三个关节的角度如下：
    髋关节：与机身圆周接触点的切线方向垂直；
    膝关节：与机身水平方向平行；
    踝关节：使两段肢体成90度，足末端与地面垂直。
  7. 如权利要求6所述的方法，其特征在于，所述髋关节能沿着机身圆周运动，使机身转动。
  8. 如权利要求6所述的方法，其特征在于，各关节旋转的预设角度的计算方法为：在每个足坐标系中，将机器人各个关节角度与足末端点坐标进行映射，通过坐标映射计算出各个足末端在其平移路径竖直平面上的运动轨迹，并获得足末端在该运动轨迹不同坐标点上对应的各关节旋转角度，配置关节旋转预设角度；
    足末端在运动轨迹不同坐标点上对应的各关节旋转角度采取D-H矩阵进行计算，具体公式为：足末端的变换矩阵T为：[T]＝[Z1][Z2]...[Zn]；
    相邻两连杆间的变换矩阵Zi为：
    Figure PCTCN2017081302-appb-100002
    其中，θi表示运动时各关节绕自身坐标系z轴的旋转角；di表示运动时在z轴上各关节之间的偏移；i取值为1至3。
  9. 如权利要求1所述的方法，其特征在于，每足按相同周期、不同相位差进行循环动作，各足平稳交替平移；
    每足的运动周期大致分为三部分：抬起、下落、后蹬；其中，后蹬部分时间占到周期的2/3。
  10. 如权利要求9所述的方法，其特征在于，每足的运动周期的三部分动作的实现是通过将每个运动周期足末端的运动轨迹均匀设置多个关键坐标点，再根据关键坐标点数量将整个运动周期分为均匀的小周期，然后在一个小周期内将足末端从一个关键坐标点移到下一个关键坐标点。
PCT/CN2017/081302 2016-04-21 2017-04-20 提高六足机器人行走稳定性的步态规划方法 WO2017181976A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610251469.1 2016-04-21
CN201610251469.1A CN105666498B (zh) 2016-04-21 2016-04-21 提高六足机器人行走稳定性的步态规划方法

Publications (1)

Publication Number Publication Date
WO2017181976A1 true WO2017181976A1 (zh) 2017-10-26

Family

ID=56215846

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/081302 WO2017181976A1 (zh) 2016-04-21 2017-04-20 提高六足机器人行走稳定性的步态规划方法

Country Status (2)

Country Link
CN (1) CN105666498B (zh)
WO (1) WO2017181976A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107933736A (zh) * 2017-12-07 2018-04-20 程子铭 多足机器人
CN109635498A (zh) * 2018-12-29 2019-04-16 中国科学院合肥物质科学研究院 一种基于接触交互信息的多足机器人行走稳定性评价方法
CN109911241A (zh) * 2019-03-20 2019-06-21 成都飞机工业（集团）有限责任公司 一种基于七次多项式的多部段自动化调姿的调姿方法
CN111913490A (zh) * 2020-08-18 2020-11-10 山东大学 基于落足调整的四足机器人动步态稳定控制方法及系统
CN111949929A (zh) * 2020-08-12 2020-11-17 智能移动机器人(中山)研究院 一种多传感器融合的四足机器人运动里程计设计方法
CN113985874A (zh) * 2021-10-26 2022-01-28 西北工业大学 一种基于CPG-Hopf网络耦合算法的水下六足机器人步态生成与转换方法
CN114721414A (zh) * 2022-04-08 2022-07-08 合肥工业大学 一种四足机器人基于trot步态的平衡控制算法
CN116719335A (zh) * 2023-06-07 2023-09-08 哈尔滨理工大学 一种考虑随机腿部故障的六足机器人容错步态规划方法
CN117067223A (zh) * 2023-10-16 2023-11-17 哈尔滨理工大学 一种基于运动稳定性估计的六足机器人自由步态规划方法
CN117242958A (zh) * 2023-11-16 2023-12-19 龙门实验室 一种适用于温室的移栽机器人

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105666498B (zh) * 2016-04-21 2017-07-07 奇弩（北京）科技有限公司 提高六足机器人行走稳定性的步态规划方法
CN109955928B (zh) * 2017-12-25 2020-10-16 深圳市优必选科技有限公司 一种双足机器人及其等效轨迹生成方法和装置
CN108500982A (zh) * 2018-04-04 2018-09-07 深圳市云电机器人技术有限公司 仿生多足机器人运动模式的构建方法及其系统
US11560192B2 (en) * 2020-04-24 2023-01-24 Ubtech Robotics Corp Ltd Stair climbing gait planning method and apparatus and robot using the same

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6481513B2 (en) * 2000-03-16 2002-11-19 Mcgill University Single actuator per leg robotic hexapod
CN102749919A (zh) * 2012-06-15 2012-10-24 华中科技大学 一种多足机器人平衡控制方法
CN104192221A (zh) * 2014-09-26 2014-12-10 哈尔滨工业大学 一种电驱动六足机器人运动控制系统及方法
CN105172933A (zh) * 2015-08-18 2015-12-23 长安大学 一种仿蜘蛛的多足机器人平台
CN105666498A (zh) * 2016-04-21 2016-06-15 奇弩（北京）科技有限公司 提高六足机器人行走稳定性的步态规划方法
CN105773618A (zh) * 2016-04-21 2016-07-20 奇弩（北京）科技有限公司 一种六足机器人行走的实现方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN203793468U (zh) * 2014-04-24 2014-08-27 中北大学 全地形水平运载机器人
CN204173037U (zh) * 2014-07-08 2015-02-25 湖北第二师范学院 一种六足机器人

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6481513B2 (en) * 2000-03-16 2002-11-19 Mcgill University Single actuator per leg robotic hexapod
CN102749919A (zh) * 2012-06-15 2012-10-24 华中科技大学 一种多足机器人平衡控制方法
CN104192221A (zh) * 2014-09-26 2014-12-10 哈尔滨工业大学 一种电驱动六足机器人运动控制系统及方法
CN105172933A (zh) * 2015-08-18 2015-12-23 长安大学 一种仿蜘蛛的多足机器人平台
CN105666498A (zh) * 2016-04-21 2016-06-15 奇弩（北京）科技有限公司 提高六足机器人行走稳定性的步态规划方法
CN105773618A (zh) * 2016-04-21 2016-07-20 奇弩（北京）科技有限公司 一种六足机器人行走的实现方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
KANG, YIN'GE: "Design and Simulation of a Hydraulic Hexapod Robot", CHINA MASTER'S THESES FULL-TEXT DATABASE, 15 September 2014 (2014-09-15), pages 24 - 25, ISSN: 1674-2046 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107933736A (zh) * 2017-12-07 2018-04-20 程子铭 多足机器人
CN109635498A (zh) * 2018-12-29 2019-04-16 中国科学院合肥物质科学研究院 一种基于接触交互信息的多足机器人行走稳定性评价方法
CN109635498B (zh) * 2018-12-29 2023-03-24 中国科学院合肥物质科学研究院 一种基于接触交互信息的多足机器人行走稳定性评价方法
CN109911241B (zh) * 2019-03-20 2022-06-14 成都飞机工业（集团）有限责任公司 一种基于七次多项式的多部段自动化调姿的调姿方法
CN109911241A (zh) * 2019-03-20 2019-06-21 成都飞机工业（集团）有限责任公司 一种基于七次多项式的多部段自动化调姿的调姿方法
CN111949929A (zh) * 2020-08-12 2020-11-17 智能移动机器人(中山)研究院 一种多传感器融合的四足机器人运动里程计设计方法
CN111913490A (zh) * 2020-08-18 2020-11-10 山东大学 基于落足调整的四足机器人动步态稳定控制方法及系统
CN111913490B (zh) * 2020-08-18 2023-11-14 山东大学 基于落足调整的四足机器人动步态稳定控制方法及系统
CN113985874A (zh) * 2021-10-26 2022-01-28 西北工业大学 一种基于CPG-Hopf网络耦合算法的水下六足机器人步态生成与转换方法
CN114721414A (zh) * 2022-04-08 2022-07-08 合肥工业大学 一种四足机器人基于trot步态的平衡控制算法
CN116719335A (zh) * 2023-06-07 2023-09-08 哈尔滨理工大学 一种考虑随机腿部故障的六足机器人容错步态规划方法
CN116719335B (zh) * 2023-06-07 2024-04-19 哈尔滨理工大学 一种考虑随机腿部故障的六足机器人容错步态规划方法
CN117067223A (zh) * 2023-10-16 2023-11-17 哈尔滨理工大学 一种基于运动稳定性估计的六足机器人自由步态规划方法
CN117067223B (zh) * 2023-10-16 2024-01-05 哈尔滨理工大学 一种基于运动稳定性估计的六足机器人自由步态规划方法
CN117242958A (zh) * 2023-11-16 2023-12-19 龙门实验室 一种适用于温室的移栽机器人
CN117242958B (zh) * 2023-11-16 2024-02-02 龙门实验室 一种适用于温室的移栽机器人

Also Published As

Publication number Publication date
CN105666498A (zh) 2016-06-15
CN105666498B (zh) 2017-07-07

Similar Documents

Publication Publication Date Title
WO2017181976A1 (zh) 提高六足机器人行走稳定性的步态规划方法
JP6501921B2 (ja) 二足ロボットの歩行制御方法及び歩行制御装置
CN111913490A (zh) 基于落足调整的四足机器人动步态稳定控制方法及系统
CN105773618B (zh) 一种六足机器人行走的实现方法
CN107045552B (zh) 一种基于正弦对角步态与快速查表法的四足机器人运动控制方法及控制装置
US20130079929A1 (en) Robot and control method thereof
KR20020086468A (ko) 각식 이동 로봇을 위한 동작 제어 장치 및 동작 제어방법, 및 로봇 장치
Gong et al. Bionic quadruped robot dynamic gait control strategy based on twenty degrees of freedom
CN110039544A (zh) 基于三次样条插值的仿人足球机器人步态规划
US8301303B2 (en) Robot and method of controlling the same
Shkolnik et al. Inverse kinematics for a point-foot quadruped robot with dynamic redundancy resolution
Zhang et al. Trot pattern generation for quadruped robot based on the ZMP stability margin
CN112744306A (zh) 一种基于运动学的四足爬壁机器人转身步态控制方法
CN210634664U (zh) 一种可避障蜘蛛六足机器人
CN112697149B (zh) 一种六足机器人节律步态足端轨迹的规划方法
CN102799184A (zh) 仿生机械恐龙爬行稳定性控制方法
US8498742B2 (en) Robot and method of controlling balance thereof
JP3674779B2 (ja) 脚式移動ロボットのための動作制御装置及び動作制御方法、並びにロボット装置
Tang et al. A stable trot strategy of quadruped robot based on capture point
CN105438305B (zh) 一种仿生六肢昆虫机器人的使用方法
CN114995476B (zh) 一种步行机器人坡地崎岖地形的步态生成与姿态控制方法
Lu et al. Turning in a bipedal robot
CN108516028B (zh) 一种复式四足机器人的行走控制方法
Janardhan et al. Kinematic analysis of biped robot forward jump for safe locomotion
Bhavanibhatla et al. Kinematic analysis of the legged mobile manipulator

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17785460

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17785460

Country of ref document: EP

Kind code of ref document: A1