CN116256972A - Six-foot robot man-machine instruction combination optimization method - Google Patents
Six-foot robot man-machine instruction combination optimization method Download PDFInfo
- Publication number
- CN116256972A CN116256972A CN202211508724.8A CN202211508724A CN116256972A CN 116256972 A CN116256972 A CN 116256972A CN 202211508724 A CN202211508724 A CN 202211508724A CN 116256972 A CN116256972 A CN 116256972A
- Authority
- CN
- China
- Prior art keywords
- robot
- instruction
- driving
- driver
- man
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005457 optimization Methods 0.000 title claims abstract description 10
- 241000238631 Hexapoda Species 0.000 claims abstract description 32
- 230000007704 transition Effects 0.000 claims abstract description 19
- 238000005265 energy consumption Methods 0.000 claims abstract description 15
- 230000001960 triggered effect Effects 0.000 claims abstract 2
- 238000011156 evaluation Methods 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000005021 gait Effects 0.000 claims description 4
- 238000012876 topography Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
Abstract
The invention belongs to the technical field of foot robot driving control. The invention discloses a hexapod robot man-machine instruction combination optimization method, which solves the problem of optimization distribution of a driver and a robot decision instruction in the driving process of the hexapod robot. According to the hexapod robot man-machine instruction combination optimization method, an allocation scheme between a decision instruction of a driver and a decision instruction of the hexapod robot is used as a state of a finite state machine, when constraint conditions of a stability margin, energy consumption and collision times of the hexapod robot in a running process exceed a set threshold value, state transition is triggered, and transition probability is increased in the state transition process to improve flexibility of state transition. The method can acquire the optimal driving instruction combination between the driver and the robot in real time, and effectively improves the running performance of the hexapod robot and the adaptability to emergency conditions.
Description
Technical Field
The invention belongs to the technical field of foot robot driving operation, and particularly relates to a collaborative control method capable of efficiently and rapidly combining and optimizing driving instructions of people and machines.
Background
The foot type mobile robot has stronger adaptability and trafficability to the terrain environment compared with the wheeled or crawler type mobile robot by virtue of rich gait and flexible movement modes when facing complex terrain, the foot type mobile robot has wide application in the fields of space detection, disaster relief and rescue, weaponry and the like due to the characteristics of the foot type robot, and also has a plurality of burdens on driving operation of a driver due to complex movement parameters to be controlled by multiple degrees of freedom, and the driver has to make real-time decisions on external environment changes and self-states of a machine body in the driving process, so that the driver can certainly operate improperly, the accuracy of decision is affected, and a good running strategy cannot be obtained by solely depending on an autonomous algorithm of the robot when facing complex and changeable environments, besides the robot is controlled fully automatically or manually, the robot can also work with people with different auxiliary and automatic levels, and the man-machine cooperative decision can play respective advantages and make up for the advantages of the people and the machine, so that how to carry out reasonable man-machine cooperative work ensures that the foot type robot has a stable running state to be solved.
The finite state machine is a mathematical calculation model for representing finite states and transfer and action behaviors among the states, has wide application in the fields of computer science, circuit system design, software engineering and the like, is applied to the aspect of autonomous decision of unmanned vehicles at present due to the characteristics of limited states and flexible conversion among the states, and is often used for determining specific work division of people and machines in the aspect of man-machine cooperative work, and the contents of division of work in different occasions are also changed in real time, so that the characteristics of flexible conversion of the finite state machine are brought into play in consideration of how the finite state machine is applied to the field of man-machine cooperative work, and the best division of work can be determined in real time between man-machines, so that the problem of man-machine cooperative work is solved.
Disclosure of Invention
The invention aims to provide a hexapod robot man-machine instruction combination optimization method, which can realize optimizing distribution of driving instructions between a driver and the hexapod robot, so that the hexapod robot has a better driving effect.
The method comprises the steps of recording instructions of a driver and recording driving instructions of the hexapod robot under the position coordinates of specific terrains; the robot reproduces the instruction of the driver, and obtains the driving instruction information of the driver at the position according to the current position coordinate of the hexapod robot so as to realize autonomous decision; establishing an offline evaluation function to determine with which driving scheme combination the hexapod robot should travel in the current terrain environment; the on-line real-time man-machine driving instruction scheme distribution of the hexapod robot is realized by utilizing the finite state machine, so that the driving scheme combination can be timely adjusted when a driver or the hexapod robot has improper driving behaviors.
The invention is realized by the following technical scheme:
the six-foot robot man-machine instruction combination optimization method specifically comprises the following steps:
step 1: the method comprises the steps of (1) recording a driver instruction, wherein as shown in fig. 1, the driver makes a decision according to current environment information, and meanwhile, the gait, stride, step height, speed and pitch angle information of the hexapod robot and position coordinate information of the hexapod robot during instruction recording are subjected to data persistence;
step 2: the robot reproduces the instruction of the driver, as shown in fig. 1, the hexapod robot obtains the driving instruction information of gait, stride, step height, speed and pitch angle of the driver at the position in the step 1 according to the information of the current terrain and the current position coordinate, so as to realize autonomous decision;
step 3: establishing an offline evaluation function to determine which driving scheme combination the hexapod robot should travel in the current terrain environment, taking the stability margin, energy consumption and collision times of the hexapod robot as constraint conditions, determining model parameters by using a Gaussian kernel formula, enumerating the stability margin, energy consumption and collision times data of the hexapod robot traveling under all driving scheme combinations, calculating the score of the driving scheme combination by using the offline evaluation function, and finally determining which man-machine driving instruction distribution mode the hexapod robot will travel according to the score;
step 4: the on-line real-time man-machine driving instruction scheme distribution of the hexapod robot is realized by utilizing the finite state machine, so that the driving scheme combination can be timely adjusted when a driver or the hexapod robot has improper driving behaviors. The allocation scheme between the decision instruction of the driver and the decision instruction of the hexapod robot is used as the state of the finite state machine, when the constraint condition exceeds the set threshold value to be the state triggering condition, the state transition process is increased by the transition probability, so that the flexibility of state transition is improved, and the adaptability of the hexapod robot to the complex environment is improved.
The beneficial effects of the invention are as follows:
according to the six-legged robot man-machine instruction combination optimization method, on one hand, the problem of decision conflict when both man-machine parties make decisions can be realized, and the respective decision results are optimized and combined to form a new decision result; on the other hand, the collaborative driving control method can lead the man-machine side to give the driving authority to the party when the driving is lost, thereby ensuring the driving fault tolerance; finally, the optimal instruction distribution of the man-machine can be realized, and the running performance of the hexapod robot is improved.
Drawings
FIG. 1 is a schematic diagram of a driver instruction recording and reproducing structure
FIG. 2 is an offline decision flow chart
FIG. 3 is an online decision flow chart
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings and examples.
One embodiment of the invention: a six-foot robot man-machine instruction combination optimization method comprises the following steps:
(1): by analyzing the topographical features and geometric features of the topography, the common irregular topography can be generalized into four typical features: flat terrain, sloping field terrain, obstacle terrain, and gully terrain under which a driver commonly uses driving instructions including: when describing the terrains, the method of describing the grid map is adopted, the grid map information is divided into a plurality of layers, one layer is used for recording the height information of the terrains, the other layers are used for recording the instruction information of a driver, the association of the terrains and the driver instructions is realized, the adoption recording frequency of each instruction is reasonably set, the driving instruction using frequencies under the 4 terrains are summarized according to the driving experience of the driver, the parameters of the same terrains are further divided, the distance of the flat terrains, the gradient of the sloping terrains, the height of the obstacle terrains and the width of the gully terrains are divided into three grades of large, medium and small, the using frequency of the driver operating instructions under different terrains is divided into three grades of high, medium and low, and therefore the recording frequency of the driving instructions with high using frequency can be improved under a certain terrains, and the recording frequency of the driving instructions with low using frequency is reduced.
(2): according to the corresponding relation between the topography recorded by the driver and the instruction, the complex autonomous decision process of the robot can be simplified into a reproduction process of the recorded instruction, when the robot runs in the topography environment recorded by the driver, the driving instruction recorded in the corresponding grid can be obtained according to the current position, and the robot uses the driving instruction as a decision result of the robot.
(3): FIG. 2 is a flow chart of offline decision, in which stability margin, energy consumption and collision number data of different decision instruction combinations under the terrain are first obtained, and since 3 constraint conditions are different in unit and cannot be calculated subsequently, normalization processing is required, and then the stability margin R is used st Energy consumption R en And number of collisions R co Establishing an evaluation function E c ＝aR st +bR en +cR co The weight parameters corresponding to the constraint conditions are a, b and c, the size relation of the evaluation function parameters can be determined according to different terrain environments, the robot hardly collides when driving on a flat ground, and the fluctuation of stability is small, so that the consumption of energy is mainly considered on the flat ground, the weight of the stability is inferior, and the weight of the collision is the least; when driving on rough obstacle terrain, mainly considering collision with the obstacle, secondly stability and thirdly energy consumption; when the robot runs on a sloping field, the posture of the robot body changes when the robot goes up and down the sloping field, so that the requirement on stability is highest, the energy consumption is the energy consumption, and the collision times are the first time; when the robot passes through the gully, the stability requirement is maximum because the pose and the foot falling position of the robot body are to be adjusted, the collision problem of the shank is considered, and finally the energy consumption is considered, and three constraint conditions are considered to be specificThe weight values of the type of terrain have obvious magnitude relation, the weight values of the type of terrain are also related to the terrain parameters, and the Gaussian kernel function is utilized to determine the weight values of the constraint function
The magnitude relation of the weight of the constraint function under the terrain can be expressed according to the distances from the constraint condition under different terrains to the central position, the specific weight value of the constraint function is related to the magnitude of sigma, the sigma can be determined by using the parameters of the terrain, and the larger the slope of the terrains is, the smaller the sigma value is; the longer the distance to be travelled under flat terrain, the smaller the value of sigma; the greater the number of obstacles under the obstacle topography, the smaller the value of sigma; the wider the width of the trench σ, the smaller the value of the trench.
(4) In the method, in order to make a decision process more real-time and efficient, a finite state machine method is used to realize dynamic decision between man-machine driving instructions, as shown in fig. 3, an online decision flow chart is shown, states of the state machine are distribution combinations of driving instructions, when a constraint condition of a robot body is changed to an unfavorable direction due to a parameter controlled by a driver or the robot, a control authority of the parameter is required to be removed, namely, state transition occurs, a transition condition between states is related to the constraint condition of the robot, specifically, the current stability margin is too low, energy consumption is too high, collision times are all triggering conditions of state transition, in order to make the transition process more flexible, the probability of transition is transferred from the fixed probability, the probability of occurrence under the current driving instruction can be represented by the driving instruction ratio, the probability of occurrence between the current state and the next state is simplified according to the state transition probability, and the current state transition probability can be represented by the driving instruction under the independent state, and the probability of the driving instruction can be obtained under the condition that the current state transition probability is represented by the driving instruction, and the driving instruction can be represented by the driving instruction is obtained under the condition of the following the driving instruction:
event a represents the current state and event B represents the next state. />
Claims (3)
1. The six-foot robot man-machine instruction combination optimization method is characterized by comprising the following steps of:
step 1: the method comprises the steps of recording instructions of a driver, recording the instructions of the driver of a six-foot robot in the positions of specific training terrains (wherein the specific training terrains refer to regular obstacle terrains such as sloping lands, step ravines and the like matched with the dimensions of the robot, coordinates of foot ends and mass centers of an organism in a three-dimensional fixed coordinate system), and aiming at different using frequencies of different instructions of the driver under different training terrains, dynamically changing the recording frequencies of the different instructions according to the training terrains, so that the using frequencies and the recording frequencies of the instructions are in positive correlation;
step 2: the robot reproduces the instruction of the driver, and according to the current position coordinates of the six-foot robot, the driving instruction sequences of the gait, the stride, the step height, the speed and the pitch angle of the driver at the position in the step 1 are fetched, and the group of instruction sequences are automatically executed to realize the autonomous decision of the robot;
step 3: according to the combined data of the man-machine driving instructions acquired offline under different training terrains, evaluating the preselected combination of the man-machine driving instructions entering the different training terrains by combining an offline decision-making evaluation function;
step 4: the conversion of the man-machine driving instruction combination is realized by using a finite state machine under different states, and the optimal man-machine instruction combination under the terrain is dynamically selected.
2. An offline decision-making evaluation function according to claim 1, step 3, characterized in that: taking the stability margin, the energy consumption and the collision times of the hexapod robot as constraint conditions for evaluating the running performance of the hexapod robot, and establishing an offline evaluation function:
e=arst+brn+crco (1) where E represents an offline evaluation score, and Rst, ren, rco is a result obtained by normalizing and unifying correlations of a stability margin, energy consumption and collision times of the hexapod robot, respectively, in consideration of differences of magnitude and correlations of three evaluation dimensions, and a, b and c are weight parameters corresponding to the stability margin, the energy consumption and the collision times, respectively, and since weights of constraint conditions change with different training terrains, a gaussian kernel function is used to calculate the weight parameters:
the process of the weight parameter can be classified into: taking a cluster of straight lines parallel to the radial central axis of the Gaussian kernel function, intersecting the Gaussian function curve, wherein the longitudinal coordinate value of the intersection point is a weight value, x is expressed as the distance between constraint conditions and the central axis, sigma is used for expressing the magnitude grade of the terrain parameter, the steeper the grade is, the steeper the terrain is, and the magnitude of the weight parameter expressed by y can be determined according to the values of x and sigma;
the weight of the three evaluation indexes is determined according to the type difference of the terrain, and then the values of the six-foot robot constraint conditions under different instruction combinations are brought into the evaluation function, so that the instruction combination with the highest offline evaluation score can be obtained to enter the terrain.
3. The method for realizing the combined conversion of the man-machine driving instructions of the hexapod robot in different states by using the finite state machine according to the step 4 of the claim 1 is characterized in that: taking a combination of driving instructions of a driver and the six-legged robot as a state of a finite state machine; the transition condition between the states is related to the constraint condition of the offline function in the step 3, and when the stability margin of the hexapod robot is too low, the energy consumption is too high and the collision times are too high in the running process, the state transition is triggered; and taking the duty ratio of the driving scheme combination as the probability of the driving scheme combination according to the constraint condition of each driving scheme combination in the step 3:
pst in i 、Pen i 、Pco i To generate a stability margin energy consumption and probability of each state under the crash frequency conversion condition, it is equal to the value Rst of the constraint condition under each state i 、Ren i 、Rco i And the ratio of the total constraint condition to the total constraint condition in each state is n, and the number of states is considered to be mutually independent, so that the probability of occurrence of the driving scheme combination is the probability of transition between the driving scheme combinations, thereby improving the flexibility of state transition and enabling the hexapod robot to have stronger adaptability to emergency conditions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211508724.8A CN116256972B (en) | 2022-11-29 | 2022-11-29 | Six-foot robot man-machine instruction combination optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211508724.8A CN116256972B (en) | 2022-11-29 | 2022-11-29 | Six-foot robot man-machine instruction combination optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116256972A true CN116256972A (en) | 2023-06-13 |
| CN116256972B CN116256972B (en) | 2024-01-23 |
Family
ID=86686931
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211508724.8A Active CN116256972B (en) | 2022-11-29 | 2022-11-29 | Six-foot robot man-machine instruction combination optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116256972B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110172825A1 (en) * | 2010-01-12 | 2011-07-14 | Samsung Electronics Co., Ltd. | Walking control apparatus of robot and method of controlling the same |
| CN104331075A (en) * | 2014-09-28 | 2015-02-04 | 哈尔滨工业大学 | Six-foot robot control system and control method thereof |
| CN108400937A (en) * | 2018-02-23 | 2018-08-14 | 北京交通大学 | The method for routing of underground coal mine wireless multimedia sensor network Differentiated Services |
| CN112571418A (en) * | 2020-12-15 | 2021-03-30 | 山东科技大学 | Four-footed robot motion state clustering and autonomous decision-making method |
| CN113394817A (en) * | 2021-06-28 | 2021-09-14 | 国网甘肃省电力公司电力科学研究院 | Multi-energy capacity optimal configuration method of wind, light, water and fire storage system |
| CN114834485A (en) * | 2022-05-29 | 2022-08-02 | 哈尔滨理工大学 | Man-machine cooperative control system and method for manned foot type mobile platform |
-
2022
- 2022-11-29 CN CN202211508724.8A patent/CN116256972B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110172825A1 (en) * | 2010-01-12 | 2011-07-14 | Samsung Electronics Co., Ltd. | Walking control apparatus of robot and method of controlling the same |
| CN104331075A (en) * | 2014-09-28 | 2015-02-04 | 哈尔滨工业大学 | Six-foot robot control system and control method thereof |
| CN108400937A (en) * | 2018-02-23 | 2018-08-14 | 北京交通大学 | The method for routing of underground coal mine wireless multimedia sensor network Differentiated Services |
| CN112571418A (en) * | 2020-12-15 | 2021-03-30 | 山东科技大学 | Four-footed robot motion state clustering and autonomous decision-making method |
| CN113394817A (en) * | 2021-06-28 | 2021-09-14 | 国网甘肃省电力公司电力科学研究院 | Multi-energy capacity optimal configuration method of wind, light, water and fire storage system |
| CN114834485A (en) * | 2022-05-29 | 2022-08-02 | 哈尔滨理工大学 | Man-machine cooperative control system and method for manned foot type mobile platform |
Non-Patent Citations (1)
| Title |
|---|
| 金马: "基于Vortex平台的重载六足机器人动力学仿真研究", 哈尔滨工业大学硕士学位论文, pages 1 - 3 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116256972B (en) | 2024-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110989576A (en) | Target following and dynamic obstacle avoidance control method for differential slip steering vehicle | |
| CN110716562A (en) | Decision-making method for multi-lane driving of unmanned vehicle based on reinforcement learning | |
| CN107357168A (en) | A kind of unmanned vehicle barrier-avoiding method based on Chance-constrained Model PREDICTIVE CONTROL | |
| CN112327888B (en) | Path planning method and device, electronic equipment and storage medium | |
| CN101916113A (en) | Automotive body gesture decoupling control method based on active suspension evaluation indicator | |
| Li et al. | Dynamically integrated spatiotemporal‐based trajectory planning and control for autonomous vehicles | |
| He et al. | Robust multiagent reinforcement learning toward coordinated decision-making of automated vehicles | |
| US11911902B2 (en) | Method for obstacle avoidance in degraded environments of robots based on intrinsic plasticity of SNN | |
| CN110514224B (en) | Method for evaluating local path planning performance of unmanned vehicle | |
| CN113859226B (en) | Motion planning and automatic parking method based on reinforcement learning | |
| Wu et al. | Trajectory prediction based on planning method considering collision risk | |
| Gu et al. | Genetic algorithm based lqr control for agv path tracking problem | |
| CN116256972B (en) | Six-foot robot man-machine instruction combination optimization method | |
| CN114384916A (en) | Adaptive decision-making method and system for off-road vehicle path planning | |
| CN117007066A (en) | Unmanned trajectory planning method integrated by multiple planning algorithms and related device | |
| CN114537435A (en) | Real-time whole vehicle track planning method in automatic driving | |
| CN115214672A (en) | Automatic driving type human decision-making, planning and controlling method considering workshop interaction | |
| CN113959446A (en) | Robot autonomous logistics transportation navigation method based on neural network | |
| Hu et al. | Real-time evaluation algorithm of human body movement in football training robot | |
| Zeng et al. | $ DBO $ Trajectory Planning and $ HAHP $ Decision-Making for Autonomous Vehicle Driving on Urban Environment | |
| Bai et al. | Some rules for setting the horizon parameters of NMPC-based vehicle path tracking | |
| Öztürk et al. | A new speed planning method based on predictive curvature calculation for autonomous driving | |
| Hou et al. | GA-based velocity planning using jerk as the encoding method for autonomous vehicles | |
| CN111857112A (en) | Automobile local path planning method and electronic equipment | |
| Gao et al. | Balancing accuracy and efficiency: Fast motion planning based on nonlinear model predictive control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |