CN110281228A - A kind of anthropomorphic robot crosses the planning control method of vertical obstacle - Google Patents

Abstract

The invention discloses a planning control method for a humanoid robot to cross a vertical barrier, formulate the target object that the humanoid robot needs to cross; perform pre-planning of the motion track of the humanoid robot according to the target object; according to the pre-planned motion track Calculate the expected take-off speed of the humanoid robot; according to the expected take-off speed, apply motor characteristic curve constraints, joint limit constraints and ground friction constraints on the hip joint, knee joint and ankle joint of the humanoid robot respectively to obtain the actual take-off speed, and then optimize The trajectory of the take-off phase; in the air phase, the stability of the upper body of the humanoid robot is maintained through the stable adjustment of the attitude in the air; in the landing phase, the stable adjustment of the landing attitude is used to slow down the landing impact, thereby completing the entire task of crossing the vertical barrier. The invention realizes a larger jumping distance without an energy storage element by optimizing the joint force moment, surmounts vertical barriers, and maintains the stability of the humanoid robot itself during the process.

Description

A planning control method for a humanoid robot to cross vertical barriers

technical field

The invention belongs to the technical field of robots, and in particular relates to a planning control method for a humanoid robot to cross a vertical barrier.

Background technique

Compared with other forms of robots such as wheeled and tracked robots, humanoid robots have stronger mobility in complex environments due to their unique bipedal movement. This advantage of the humanoid robot makes it have a strong advantage in special environments such as dangerous environments and rescue and disaster relief. In the face of most flat and continuous road environments, humanoid robots can adapt by crawling or walking. However, when a humanoid robot encounters discontinuous road surfaces, such as complex and harsh terrain environments such as ditches and barriers, it needs to be applied to a movement mode similar to humans-forward jumping.

A kangaroo-like jumping robot is proposed in the prior art, and a ratchet device is proposed to control the energy accumulation and release of the jumping mechanism to realize the jumping action. However, other motion modes such as walking or crawling of the robot are not considered.

In addition, there is a small humanoid robot that can use the CPG method to generate jumping trajectories for vertical jumping. However, the robot is small in size, and when the method is applied to a large robot, it may not be applicable due to insufficient driving force. A method for vertical jumping of a humanoid robot. However, the robot can only jump vertically upwards and cannot cross vertical barriers. Moreover, this paper does not consider the constraints of the motor torque in practical applications. A humanoid robot that uses compressed springs to simulate leg motions, which can jump through the resonance of pelvic motion and joint elasticity.

To sum up, most of the existing humanoid robots rely on elastic energy storage elements to achieve jumping functions, and adding elastic energy storage elements to the legs of humanoid robots often easily affects the realization of other motion modes of the robot, such as walking and crawling, resulting in robot motion. Simplification of patterns.

Contents of the invention

In order to solve the deficiencies in the prior art, the present invention proposes a planning control method for a humanoid robot to cross a vertical barrier. By optimizing the joint force moment, a larger jumping distance can be achieved without an energy storage element. Overcome vertical barriers while maintaining the stability of the humanoid robot itself.

Technical purpose of the present invention is achieved through the following technical solutions:

A planning control method for a humanoid robot to cross a vertical barrier, comprising the following steps:

Step 1, formulate the target object that the humanoid robot needs to cross; for the target object, perform pre-planning of the trajectory of the humanoid robot;

Step 2, calculate the expected take-off speed of the humanoid robot according to the pre-planned trajectory;

Step 3. According to the expected take-off speed, the humanoid robot hip joint, knee joint and ankle joint are subjected to motor characteristic curve constraints, joint limit constraints and ground friction constraints respectively, so as to obtain the actual take-off speed, and then optimize the trajectory of the take-off phase ;

Step 4, maintain the stability of the upper body of the humanoid robot through the stable adjustment of the attitude in the air during the flight phase;

Step 5, during the landing stage, the impact of landing is slowed down by stabilizing the landing posture adjustment, so as to complete the entire task of crossing the vertical barrier.

Further, the method for stabilizing the attitude in the air is as follows:

By adjusting the ankle joint of the robot, the feet of the robot are always kept parallel to the ground to prepare for the landing of the robot; set the height threshold between the robot and the ground, and when the height of the descending stage reaches the threshold, the hip and knee joints are changed The speed of the center of mass of the robot relative to the soles of the feet is used to reduce the speed of the soles of the robot relative to the ground, making it tend to 0, thereby reducing the landing impact of the robot.

Further, the method for stabilizing the landing attitude is:

The hip joint, knee joint and ankle joint of the humanoid robot are constrained by the motor characteristic curve, the joint limit constraint and the ground friction force constraint respectively, so that the initial position of the joint is restored while the speed of each joint tends to zero, that is, the robot returns to the initial standing posture;

Further, the motor torque characteristic curve is constrained as:

Among them, τ _joint represents the joint torque, τ _peak is the turning point torque of the motor, is the joint angular velocity, is the joint angular velocity at the turning point, i is the motor reduction ratio, ω _max is the maximum motor speed, ω _break is the motor speed at the turning point, is the maximum joint angular velocity; the specific calculation is:

Further, the joint limit constraints are:

-90°< _q1 <0;

0< _q2 <180°;

-180°< _q3 <0;

Among them, q ₁ , q ₂ , and q ₃ respectively represent the angles of the ankle joint, knee joint and hip joint of the humanoid robot;

Further, the ground friction constraint is: to ensure that the angle between the ground force _FR and the vertical direction of the humanoid robot must be smaller than the friction angle Ensure that there is no relative movement between the soles of the robot and the ground;

Beneficial effects of the present invention:

In the absence of elastic energy storage elements, the robot's take-off trajectory and landing trajectory are optimized by combining the motor torque-speed characteristic curve to ensure that the robot's joints can reach the desired position and speed efficiently and stably. At the same time, in the air stage of the robot, the robot's posture is stable and the landing is smooth by adjusting the joints of the whole body. While achieving efficient, stable and complete forward jumping of the robot, no elastic elements are added to affect the normal progress of other movement modes of the robot such as walking and crawling.

Description of drawings

Fig. 1 is a schematic diagram of the structure of a humanoid robot;

Fig. 2 is a motor characteristic curve;

Fig. 3 is a schematic diagram of the process stages of the whole jump of the humanoid robot;

Fig. 4 is the constraint figure of ground friction force when the present invention lands;

Fig. 5 is a control flow diagram of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

A humanoid robot has three parts, a torso, a thigh and a lower leg, which are connected by hip joints, knee joints and ankle joints respectively. Since the body is always symmetrical during jumping, the structure of the humanoid robot is simplified from a three-dimensional space to a two-dimensional sagittal plane, as shown in Figure 1.

As shown in Figures 3 and 5, a kind of planning control method that the present invention designs is used for the humanoid robot to cross the vertical barrier, comprises the following steps:

Step 1. Formulate the target object that the humanoid robot needs to cross; for the target object, pre-plan the trajectory of the humanoid robot, that is, plan the distance that the humanoid robot needs to cross and the maximum high.

Step 2, calculate the expected take-off velocity v _{jump of the humanoid robot according to the pre-planned trajectory, d} = (v _{x, d} , v _{y, d} );

Step 3, taking the jumping speed v _{jump, d} = (v _{x, d} , v _{y, d} ) as the optimization goal, respectively carry out motor characteristic curve constraints, joint limit constraints and Ground friction constraints, so as to obtain the actual take-off velocity v _{jump, r} = (v _{x, r} , v _{y, r} ), and compare the actual take-off velocity v _{jump, r} = (v _{x, r} , v _{y, r} ) with the expected The difference between the speed v _{jump and} d is used as the optimization target to establish a cost function, and the pre-planned motion trajectory of the humanoid robot is optimized through the cost function; the constructed cost function is expressed as:

min J＝|v _x，r -v _x，d |+|v _y，r -v _y，d | (1)

Among them, J is the cost function, v _{x, r} is the horizontal component of the actual take-off velocity, v _{x, d} is the horizontal component of the expected take-off velocity, v _{y, r} is the vertical component of the actual take-off velocity, v _{y, d} is the expected Vertical component of take-off velocity.

Step 4, when the humanoid robot enters the vacating stage shown in Figure 3, stabilize the attitude of the robot in the air. Since the ankle joint of the robot is underactuated during the vacating phase, firstly, by adjusting the ankle joint of the robot, the foot of the robot is always parallel to the ground to prepare for the landing of the robot; secondly, set the height threshold between the robot and the ground, Once the threshold is reached in the falling phase, that is, when the robot is about to land, the speed of the robot's center of mass relative to the sole is reduced by adjusting the hip joint and knee joint to reduce the speed of the robot's sole relative to the ground, making it tend to 0, thereby reducing The landing impact of the small robot maintains the stability of the upper body of the humanoid robot during the flight phase.

Step 5, when the robot enters the landing stage, it is necessary to stabilize the landing attitude of the robot. The hip joint, knee joint and ankle joint of the humanoid robot are constrained by the motor characteristic curve, the joint limit constraint and the ground friction force constraint respectively, so that the initial position of the joint is restored while the speed of each joint tends to 0, that is, the robot returns to the initial standing posture. Thereby complete the whole task of crossing the vertical barrier.

In the present invention, the take-off of step 3 and the landing stage of step 5 of the humanoid robot hip joint, knee joint and ankle joint all adopt the motor characteristic curve constraint, and the motor characteristic curve constraint is shown in Figure 2, and the specific relationship is expressed as:

Among them, τ _j o _int represents the joint torque, τ _peak is the turning point torque of the motor, is the joint angular velocity, is the joint angular velocity at the turning point, i is the motor reduction ratio, ω _max is the maximum motor speed, ω _break is the motor speed at the turning point, is the maximum joint angular velocity; the specific calculation is:

The joint limit constraints are:

-90°< _q1 <0;

0< _q2 <180°;

-180°<q3<0; ( ₃ )

Among them, (q ₁ , q ₂ , q ₃ ) respectively represent the angles of the ankle joint, knee joint and hip joint of the humanoid robot.

The ground friction constraint is shown in Figure 4. The humanoid robot is subject to the ground force F _R during the take-off phase and the landing phase of the jumping motion, which can be decomposed into the vertical support force F _N and the ground friction force F on it. _f . Among them, F _f,max represents the maximum static friction that the ground can provide, Indicates the friction angle. In order to prevent relative movement between the soles of the robot and the ground, the angle between the ground force on the robot and the vertical direction must be smaller than the friction angle

To sum up, the technical solution adopted in the present invention can achieve a larger jump height without energy storage elements, and the method of crossing vertical barriers can improve the robot's ability to jump without affecting other motion modes of the robot. The adaptability of continuous road surfaces increases the application occasions of humanoid robots.

The above embodiments are only used to illustrate the design concept and characteristics of the present invention, and its purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes or modifications based on the principles and design ideas disclosed in the present invention are within the protection scope of the present invention.

Claims (6)

1. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot, which comprises the following steps:

Step 1, the target object that anthropomorphic robot needs to cross is formulated；For target object, the movement of anthropomorphic robot is carried out Track pre-planning；

Step 2, according to the expectation take-off speed of the moving track calculation anthropomorphic robot of pre-planning；

Step 3, according to desired take-off speed, motor characteristic is carried out respectively to anthropomorphic robot hip joint, knee joint and ankle-joint Curve constraint, joint limit constraint and ground friction force constraint, to obtain practical take-off speed, and then optimize Take-off Stage Motion profile；

Step 4, flight phase is stablized by aerial statue and is adjusted, and keeps anthropomorphic robot upper body stability；

Step 5, the landing stage by landing attitude stabilization adjustment, lands impact to slow down, to complete entirely to cross over vertical wall The task of barrier.

2. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot according to claim 1, feature It is, the method that the aerial statue stablizes adjustment are as follows: by adjusting the ankle-joint of robot, protect the sole of robot always Maintain an equal level row and ground, prepares for the landing of robot；The height threshold for setting robot and ground, when the height of decline stage Reach the threshold value, changes robot mass center by adjusting hip joint and knee joint and reduce robot foot relative to the speed of sole The speed relative to ground is slapped, it is made to be intended to 0, to reduce the landing impact of robot.

3. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot according to claim 1, feature It is, the method for the landing attitude stabilization adjustment are as follows:

Anthropomorphic robot hip joint, knee joint and ankle-joint are carried out respectively motor characteristic curve constraint, joint limit constraint with And ground friction force constraint, so that each joint velocity tends to restore joint initial position while 0, i.e. robot restores just initial station Standing position state.

4. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot according to claim 1 or 3, special Sign is that the motor torque characteristic curve constrains are as follows:

Wherein, τ_jointIndicate joint moment, τ_peakFor the turning point torque of motor,For joint angular speed,For turning point Joint angular speed, i be decelerating through motor ratio, ω_maxFor maximum motor speed, ω_breakFor the motor speed of turning point,For Maximum joint angular speed.

5. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot according to claim 1 or 3, special Sign is that the joint limit constrains are as follows:

- 90 ° of < q₁< 0；

0 < q₂180 ° of <；

- 180 ° of < q₃< 0；

Wherein, q₁, q₂, q₃Respectively represent the ankle-joint of anthropomorphic robot, the angle of knee joint and hip joint.

6. a kind of planning control method for crossing vertical obstacle for anthropomorphic robot according to claim 1 or 3, special Sign is that the frictional ground force constrains are as follows: guarantees ground force F suffered by anthropomorphic robot_RWith the angle of vertical direction Angle of friction must be less thanIt protects between the sole and ground of robot and does not relatively move.