Disclosure of Invention
The application provides an obstacle avoidance method of a split type simulation robot with double walking modes, the split type simulation robot is composed of two legs which are detachably connected and a main body part above the two legs, the two legs are respectively and detachably and mechanically connected with the main body part, and first magnetic field generators are arranged at the joints of the main body part and the two legs respectively; the two legs are provided with built-in chips, so that after the two legs are separated from the main body part respectively, the two legs can execute operation corresponding to the control signal according to the received control signal; the method comprises the following steps:
s1, the split type simulation robot moves along a preset track in a preset striding walking mode, and meanwhile, a first area in front of the split type simulation robot is subjected to real-time environment signal detection by utilizing a preset environment detection sensor array on the split type simulation robot, so that a first environment signal sequence is obtained;
s2, inputting the first environment signal sequence into a preset obstacle analysis model for processing, so as to obtain an analysis result output by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
s3, judging whether the analysis result is that an obstacle exists;
s4, if the analysis result shows that the obstacle exists, carrying out obstacle classification processing according to a preset obstacle classification method so as to obtain a first obstacle category corresponding to the obstacle in the first area;
s5, judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not;
s6, if the first obstacle category is recorded in a preset split type obstacle avoidance list, separating the two legs from the main body part, and controlling power assemblies respectively arranged on the two legs so that the split type simulation robot can obtain an upward first momentum and a forward second momentum;
s7, sending an open command to a second magnetic field generator preset under the floor of the first area to open the second magnetic field generator and open the first magnetic field generator preset on the main body, so that the main body can pass over the first area and is supported by magnetic force without falling onto an obstacle;
s8, sending a rolling movement instruction to the two separated legs to enable the two legs to bypass the obstacle in a rolling movement mode; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
s9, when the main body part crosses the obstacle from the air and the two legs move at the same speed in the horizontal direction under the main body part and with the main body part, closing the first magnetic field generator to enable the main body part to fall on the two legs to form the split type simulation robot, and sending a closing instruction to the second magnetic field generator to complete the obstacle avoidance process of the split type simulation robot with the double walking mode.
Further, the first environment signal sequence is input into a preset obstacle analysis model for processing, so that an analysis result output by the obstacle analysis model is obtained; wherein, the analysis result is with obstacles or without obstacles, the obstacle analysis model is trained based on a neural network model and by using training data, and before step S2, the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training, the method includes:
s11, sample data of the designated data are called from a preset sample database to form a sample set, and the sample set is divided into a training set and a verification set according to a preset proportion; the sample data is composed of a signal sequence collected in advance and an artificial label corresponding to the signal sequence collected in advance;
s12, calling a preset neural network model, and inputting the training set into the neural network model for training to obtain a preliminary neural network model;
s13, verifying the preliminary neural network model by using the verification set to obtain a verification result;
s14, judging whether the verification result is that the verification is passed;
and S15, if the verification result is that the verification is passed, marking the preliminary neural network model as an obstacle analysis model.
Further, after the step S3 of determining whether the analysis result is an obstacle, the method includes:
s31, if the analysis result shows that no obstacle exists, controlling the split type simulation robot to move continuously along a preset track in a striding walking mode, and stopping the signal detection operation of the environment detection sensor array;
s32, after the split type simulation robot passes through the first area, detecting real-time environment signals in a second area in front of the split type simulation robot by using an environment detection sensor array preset on the split type simulation robot again, so as to obtain a second environment signal sequence;
s33, inputting the second environment signal sequence into a preset obstacle analysis model for processing, so as to obtain an analysis result output again by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
s34, judging whether the analysis result output again by the obstacle analysis model is an obstacle;
s35, if the analysis result shows that the obstacle exists, carrying out obstacle classification processing according to a preset obstacle classification method so as to obtain a second obstacle type corresponding to the obstacle in the second area;
s36, judging whether the second obstacle category is recorded in a preset split type obstacle avoidance list or not;
s37, if the second obstacle category is recorded in a preset split type obstacle avoidance list, separating the two legs from the main body part, and controlling power assemblies respectively arranged on the two legs so that the split type simulation robot can obtain an upward first momentum and a forward second momentum;
s38, sending an opening instruction to a second magnetic field generator preset below the floor of the second area to open the second magnetic field generator and open a first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force and cannot fall onto an obstacle when passing over the second area;
s39, sending a rolling movement instruction to the two separated legs to enable the two legs to bypass the obstacle in a rolling movement mode; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
s310, when the main body part crosses an obstacle from the air and the two leg parts move at the same speed in the horizontal direction under the main body part, closing the first magnetic field generator to enable the main body part to fall on the two leg parts to form the split type simulation robot, and sending a closing instruction to the second magnetic field generator under the floor of the second area.
Further, the step S4 of performing obstacle classification processing according to a preset obstacle classification method to obtain a first obstacle category corresponding to the obstacle in the first area includes:
s401, acquiring a first contour corresponding to an obstacle in the first area according to a preset contour acquisition method;
s402, calculating a similarity value of the designated contour and a preset standard contour according to a preset contour similarity calculation method, and judging whether the similarity value is greater than a preset similarity threshold value; the standard outline is pre-stored in a preset database, and different outlines stored in the database respectively correspond to different obstacle categories;
and S403, if the similarity value is greater than a preset similarity threshold, acquiring the obstacle category corresponding to the standard contour according to the corresponding relation between the preset contour and the obstacle category, and recording the obstacle category corresponding to the standard contour as the first obstacle category.
Further, the main body part is in a V-shaped convex shape at the joint, the leg part is in a V-shaped concave shape at the joint, and the V-shaped convex shape is matched with the V-shaped concave shape, so that the joint of the main body part and the leg part is in a V shape; the friction coefficient of the V-shaped concave front inclined surface is smaller than that of the rear inclined surface, and the power assembly is an ejection assembly; the front inclined surface refers to an inclined surface closer to the front of the split type simulation robot in the two inclined surfaces in the V-shaped concave shape, and the rear inclined surface refers to an inclined surface closer to the rear of the split type simulation robot in the two inclined surfaces in the V-shaped concave shape;
the step S6 of controlling the power assemblies respectively disposed on the two legs to make the split type simulation robot obtain the upward first momentum and the forward second momentum includes:
and S601, controlling the ejection assemblies respectively arranged on the two legs, so that the ejection assemblies give upward first momentum and forward second momentum to the main body part of the split type simulation robot by using the V-shaped concave backward inclined surface and the V-shaped convex shape of the main body part as force applying surfaces and by using the V-shaped concave forward inclined surface as a track.
Further, a track network is pre-arranged under the floor, a vehicle carrying a second magnetic field generator moves along the track network, and the track network is made of a non-magnetic material, and the track network sends an opening command to the second magnetic field generator preset under the floor of the first area to open the second magnetic field generator and open the first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force when passing over the first area and cannot fall onto an obstacle before the step S7, the method includes:
s61, sending the position information of the first area to the vehicle, and requesting the vehicle to obtain the position information of the first area,
moving along the network of tracks to under the floor of the first area.
The application provides an obstacle avoidance device of a split type simulation robot with double walking modes, wherein the split type simulation robot is composed of two legs which are detachably connected and a main body part above the two legs, the two legs are respectively and detachably and mechanically connected with the main body part, and first magnetic field generators are arranged at the joints of the main body part and the two legs respectively; the two legs are provided with built-in chips, so that after the two legs are separated from the main body part respectively, the two legs can execute operation corresponding to the control signal according to the received control signal; the device comprises:
the system comprises a first environment signal sequence acquisition unit, a second environment signal sequence acquisition unit and a control unit, wherein the first environment signal sequence acquisition unit is used for enabling the split type simulation robot to move along a preset track in a preset stepping walking mode, and meanwhile, a first area in front of the split type simulation robot is subjected to real-time environment signal detection by utilizing a preset environment detection sensor array on the split type simulation robot, so that a first environment signal sequence is obtained;
the analysis result acquisition unit is used for inputting the first environment signal sequence into a preset obstacle analysis model for processing so as to obtain an analysis result output by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
an analysis result judgment unit for judging whether the analysis result is an obstacle;
the obstacle classification unit is used for carrying out obstacle classification processing according to a preset obstacle classification method if the analysis result shows that the obstacle exists, so that a first obstacle class corresponding to the obstacle in the first area is obtained;
the split type obstacle avoidance list judging unit is used for judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not;
the simulation robot separation unit is used for separating the two leg parts from the main body part and controlling power components respectively arranged on the two leg parts if the first obstacle category is recorded in a preset split type obstacle avoidance list so as to enable the split type simulation robot to obtain an upward first momentum and a forward second momentum;
the opening instruction sending unit is used for sending an opening instruction to a second magnetic field generator preset below the floor of the first area so as to open the second magnetic field generator and open a first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force and cannot fall onto an obstacle when passing over the first area;
a rolling movement instruction sending unit, configured to send a rolling movement instruction to the two separated leg portions, so that the two leg portions bypass the obstacle in a rolling movement manner; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
and the simulation robot combination unit is used for closing the first magnetic field generator when the main body part crosses an obstacle from the air and the two legs move at the same speed in the horizontal direction under the main body part and with the main body part, so that the main body part falls on the two legs to form the split simulation robot, and sending a closing instruction to the second magnetic field generator, thereby completing the obstacle avoidance process of the split simulation robot with the double walking mode.
The present application provides a computer device comprising a memory storing a computer program and a processor implementing the steps of any of the above methods when the processor executes the computer program.
The present application provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of any of the above.
The obstacle avoiding method, the obstacle avoiding device, the computer equipment and the storage medium of the split type simulation robot with the double walking modes realize the obstacle avoiding method of the split type simulation robot by adopting the design of the split type simulation robot with the double walking modes, so that even when a narrow passage (or the cross-sectional area available for the bypassing at the side of an obstacle is smaller than that of the simulation robot) touches the obstacle, the obstacle can be efficiently avoided. Specifically, when the obstacle is touched and the obstacle is recorded in a preset split type obstacle avoidance list, the main body part of the split type simulation robot can cross over the obstacle, the legs of the split type simulation robot can cross over the obstacle in a rolling movement mode from the side of the obstacle, and the split type simulation robot is formed again after the obstacles are crossed, so that the obstacle avoidance method is high in adaptability due to the bypassing mode, and when the main body part of the split type simulation robot crosses over the obstacle, the main body part of the split type simulation robot is prevented from falling onto the obstacle in a temporary magnetic field supporting mode, and the overall safety is improved.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Referring to fig. 1, an embodiment of the present application provides an obstacle avoidance method for a split type simulation robot with a dual walking mode, where the split type simulation robot is composed of two leg portions that are detachably connected and a main body portion above the two leg portions, the two leg portions are respectively detachably and mechanically connected to the main body portion, and a first magnetic field generator is disposed at a connection point of the main body portion and the two leg portions; the two legs are provided with built-in chips, so that after the two legs are separated from the main body part respectively, the two legs can execute operation corresponding to the control signal according to the received control signal; the method comprises the following steps:
s1, the split type simulation robot moves along a preset track in a preset striding walking mode, and meanwhile, a first area in front of the split type simulation robot is subjected to real-time environment signal detection by utilizing a preset environment detection sensor array on the split type simulation robot, so that a first environment signal sequence is obtained;
s2, inputting the first environment signal sequence into a preset obstacle analysis model for processing, so as to obtain an analysis result output by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
s3, judging whether the analysis result is that an obstacle exists;
s4, if the analysis result shows that the obstacle exists, carrying out obstacle classification processing according to a preset obstacle classification method so as to obtain a first obstacle category corresponding to the obstacle in the first area;
s5, judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not;
s6, if the first obstacle category is recorded in a preset split type obstacle avoidance list, separating the two legs from the main body part, and controlling power assemblies respectively arranged on the two legs so that the split type simulation robot can obtain an upward first momentum and a forward second momentum;
s7, sending an open command to a second magnetic field generator preset under the floor of the first area to open the second magnetic field generator and open the first magnetic field generator preset on the main body, so that the main body can pass over the first area and is supported by magnetic force without falling onto an obstacle;
s8, sending a rolling movement instruction to the two separated legs to enable the two legs to bypass the obstacle in a rolling movement mode; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
s9, when the main body part crosses the obstacle from the air and the two legs move at the same speed in the horizontal direction under the main body part and with the main body part, closing the first magnetic field generator to enable the main body part to fall on the two legs to form the split type simulation robot, and sending a closing instruction to the second magnetic field generator to complete the obstacle avoidance process of the split type simulation robot with the double walking mode.
The obstacle avoidance method needs to be realized by a special simulation robot, namely, the obstacle avoidance method is realized by a split type simulation robot with a double-walking mode. The split type simulation robot with the double walking modes is characterized in that: the split type simulation robot is composed of two leg parts which are detachably connected and a main body part above the two leg parts, the two leg parts are respectively and detachably and mechanically connected with the main body part, and the main body part is provided with a first magnetic field generator; the two legs are provided with built-in chips, so that after the two legs are separated from the main body part respectively, the two legs can execute operation corresponding to the control signal according to the received control signal. Therefore, the split type simulation robot has two walking modes, namely one mode is the striding walking of a simulation person, and the striding walking of the simulation robot is the prior art, and is not described again; the other is a leg rolling movement, while the body part is moved from the air over an obstacle (this movement is a feature of the present application and will be explained in detail below in connection with specific steps). The execution subject of the present application can be any feasible subject, for example, a split type simulation robot terminal (referring to a terminal on a split type simulation robot). The first magnetic field generator arranged on the main body part can be positioned at any feasible position, and is preferably arranged at the joint of the main body part and the two leg parts respectively. The two legs are respectively and mechanically connected with the main body in a separable mode, and the mechanical connection can be realized in an electromagnetic switch (electromagnetic valve) mode, namely, the mechanical connection (normal state) is kept when the electromagnetic switch (electromagnetic valve) is closed, and the two legs are respectively and mechanically connected with the main body (separated state) when the electromagnetic switch (electromagnetic valve) is opened. The split type simulation robot can be made of any feasible material, preferably a non-magnetic material, or preferably a non-magnetic material mostly. Specifically, the proportion of the magnetic material in the split type simulation robot is less than 20%, preferably less than 10%, so that when the main body part of the split type simulation robot passes over an obstacle in the air and is supported by a magnetic field, the control of the magnetic field is more accurate, and the safety is higher.
As described in the above steps S1-S3, the split type simulation robot moves along a preset track in a preset striding walking manner, and simultaneously performs real-time environmental signal detection on a first area in front of the split type simulation robot by using a preset environmental detection sensor array on the split type simulation robot, so as to obtain a first environmental signal sequence; inputting the first environment signal sequence into a preset obstacle analysis model for processing, so as to obtain an analysis result output by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training; and judging whether the analysis result is that the obstacle exists or not. When the split type simulation robot does not encounter an obstacle, a human-simulated striding walking mode is adopted, and the action track of the split type simulation robot is a preset track. The environment detection sensor array may be any feasible sensor array, such as an image sensor array, an ultrasonic sensor array, an infrared sensor array, and/or a laser sensor array, among others. And inputting the first environment signal sequence into a preset obstacle analysis model for processing, thereby obtaining an analysis result output by the obstacle analysis model. The obstacle analysis model is a preliminary analysis to determine whether an obstacle is present.
Further, the first environment signal sequence is input into a preset obstacle analysis model for processing, so that an analysis result output by the obstacle analysis model is obtained; wherein, the analysis result is with obstacles or without obstacles, the obstacle analysis model is trained based on a neural network model and by using training data, and before step S2, the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training, the method includes:
s11, sample data of the designated data are called from a preset sample database to form a sample set, and the sample set is divided into a training set and a verification set according to a preset proportion; the sample data is composed of a signal sequence collected in advance and an artificial label corresponding to the signal sequence collected in advance;
s12, calling a preset neural network model, and inputting the training set into the neural network model for training to obtain a preliminary neural network model;
s13, verifying the preliminary neural network model by using the verification set to obtain a verification result;
s14, judging whether the verification result is that the verification is passed;
and S15, if the verification result is that the verification is passed, marking the preliminary neural network model as an obstacle analysis model.
Thereby ensuring that the obstacle analysis model is competent for the obstacle analysis task. The neural network model may be any feasible model, such as a long and short type memory model, a BP neural network model, a convolutional neural network model, or a recurrent neural network model. The preset ratio is, for example, 9: 1. The training set and the verification set are from the same batch of sample sets, so that the training and verification processes are consistent, and the reliability of the verification result is guaranteed.
Further, after the step S3 of determining whether the analysis result is an obstacle, the method includes:
s31, if the analysis result shows that no obstacle exists, controlling the split type simulation robot to move continuously along a preset track in a striding walking mode, and stopping the signal detection operation of the environment detection sensor array;
s32, after the split type simulation robot passes through the first area, detecting real-time environment signals in a second area in front of the split type simulation robot by using an environment detection sensor array preset on the split type simulation robot again, so as to obtain a second environment signal sequence;
s33, inputting the second environment signal sequence into a preset obstacle analysis model for processing, so as to obtain an analysis result output again by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
s34, judging whether the analysis result output again by the obstacle analysis model is an obstacle;
s35, if the analysis result shows that the obstacle exists, carrying out obstacle classification processing according to a preset obstacle classification method so as to obtain a second obstacle type corresponding to the obstacle in the second area;
s36, judging whether the second obstacle category is recorded in a preset split type obstacle avoidance list or not;
s37, if the second obstacle category is recorded in a preset split type obstacle avoidance list, separating the two legs from the main body part, and controlling power assemblies respectively arranged on the two legs so that the split type simulation robot can obtain an upward first momentum and a forward second momentum;
s38, sending an opening instruction to a second magnetic field generator preset below the floor of the second area to open the second magnetic field generator and open a first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force and cannot fall onto an obstacle when passing over the second area;
s39, sending a rolling movement instruction to the two separated legs to enable the two legs to bypass the obstacle in a rolling movement mode; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
s310, when the main body part crosses an obstacle from the air and the two leg parts move at the same speed in the horizontal direction under the main body part, closing the first magnetic field generator to enable the main body part to fall on the two leg parts to form the split type simulation robot, and sending a closing instruction to the second magnetic field generator under the floor of the second area.
Therefore, the split type simulation robot can walk correspondingly whether to detect the obstacle or not until reaching the terminal point. The split type simulation robot adopts an intermittent detection strategy, namely if the analysis result shows that no barrier exists, the split type simulation robot is controlled to continuously move along a preset track in a striding walking mode, and the signal detection operation of the environment detection sensor array is stopped; split type emulation robot passes through behind the first region, reuse preset environmental detection sensor array on the split type emulation robot is right the second region in split type emulation robot the place ahead carries out real-time environment signal detection to obtain second environment signal sequence. Therefore, the data calculation amount and the energy consumption are reduced, and the service life of the environment detection sensor array of the split type simulation robot is prolonged.
As described in the above steps S4-S6, if the analysis result indicates that there is an obstacle, performing obstacle classification processing according to a preset obstacle classification method, thereby obtaining a first obstacle category corresponding to the obstacle in the first area; judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not; if the first obstacle category is recorded in a preset split type obstacle avoidance list, the two leg portions are separated from the main body portion, and power assemblies respectively arranged on the two leg portions are controlled, so that the split type simulation robot can obtain an upward first momentum and a forward second momentum. Although the preliminary detection has been performed, it is determined that there is an obstacle. However, whether the obstacle avoidance operation is required is not necessarily required. For example, if a pair of shoes as obstacles is placed in the first area, it is not necessary to perform obstacle avoidance because the shoes can be strided by striding walking; and if the dining car exists in the first area, the result is the opposite. Therefore, according to the method, the obstacle classification processing is carried out according to a preset obstacle classification method, so that a first obstacle category corresponding to the obstacle in the first area is obtained; and judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not so as to determine whether obstacle avoidance operation should be performed or not. The preset obstacle classification method may adopt any feasible method, for example, the obstacle is identified by an image identification technology, so as to obtain the category of the obstacle. It should be noted that, the obstacle classification method at this time may use the data in the first environment signal sequence, or may perform signal detection again by using the environment detection sensor array to obtain a new environment signal sequence (for example, an image signal sequence), and then perform obstacle classification according to the new environment signal sequence; at this time, the environment detection sensor array has two operation modes, namely, a power saving mode (for example, only a part of the sensors operate) and an efficient mode (for example, all the sensors operate), when the first environment signal sequence is acquired, the environment detection sensor array is in the power saving mode, and when the environment detection sensor array performs signal detection again to obtain a new environment signal sequence, the environment detection sensor array is in the efficient mode. Through the design, the environment detection sensor array is not required to be in a high-efficiency mode with high power consumption and high calculation amount in the whole walking process, so that energy and electricity are saved on the premise of ensuring the identification accuracy. The split type obstacle avoidance list records the types of obstacles capable of adopting split type obstacle avoidance, such as a wooden dining car and the like. Further, the heights of the obstacles corresponding to the types of the obstacles recorded in the split-type obstacle avoidance list are all low, and more specifically, the heights of the obstacles corresponding to the types of the obstacles recorded in the split-type obstacle avoidance list are all smaller than the height of the waist of the split-type simulation robot, so that the split-type simulation robot does not need to acquire excessive upward momentum when the split-type obstacle avoidance is required. It should be noted that the obstacle to be avoided in this application may be any feasible obstacle, but is preferably an obstacle made of a non-magnetic material or an obstacle having a non-magnetic material as a main component. Moreover, the types of the obstacles in the working area of the split type simulation robot can be collected in advance, the types of the obstacles with the main components of magnetic materials are not recorded in the split type obstacle avoidance list, and the types of other obstacles are recorded in the split type obstacle avoidance list, so that the split type obstacle avoidance list can be used for screening the obstacles, and whether the split type obstacle avoidance method is adopted or not is determined. If the first obstacle category is recorded in a preset split type obstacle avoidance list, the two leg portions are separated from the main body portion, and power assemblies respectively arranged on the two leg portions are controlled, so that the split type simulation robot can obtain an upward first momentum and a forward second momentum. The two leg portions may be separated from the main body portion by any feasible method, for example, when the two leg portions are connected to the main body portion by an electromagnetic valve, the two leg portions are separated from the main body portion by opening the electromagnetic valve. The power assemblies respectively arranged on the two legs are controlled to enable the split type simulation robot to acquire the upward first momentum and the forward second momentum in any feasible mode, for example, human-simulated jumping operation is adopted, so that the split type simulation robot acquires the upward first momentum and the forward second momentum and the like.
Further, the step S4 of performing obstacle classification processing according to a preset obstacle classification method to obtain a first obstacle category corresponding to the obstacle in the first area includes:
s401, acquiring a first contour corresponding to an obstacle in the first area according to a preset contour acquisition method;
s402, calculating a similarity value of the designated contour and a preset standard contour according to a preset contour similarity calculation method, and judging whether the similarity value is greater than a preset similarity threshold value; the standard outline is pre-stored in a preset database, and different outlines stored in the database respectively correspond to different obstacle categories;
and S403, if the similarity value is greater than a preset similarity threshold, acquiring the obstacle category corresponding to the standard contour according to the corresponding relation between the preset contour and the obstacle category, and recording the obstacle category corresponding to the standard contour as the first obstacle category.
The method aims to identify whether the obstacle can pass through by adopting the obstacle avoidance method, so that the first obstacle category is obtained in a special mode, namely only the similarity between the contour acquisition and the contour is calculated, the calculated amount is reduced, and the overall identification speed is increased. The contour acquiring method may be any feasible method, for example, when the image signal of the obstacle is acquired, the contour of the image signal is processed. The contour similarity calculation method may be any feasible method, for example, overlapping two contours to overlap centers of the two contours, calculating an overlapping area and a non-overlapping area of the two contours, and then removing the non-overlapping area from the overlapping area to obtain an area ratio, where the area ratio is the similarity value.
Further, the main body part is in a V-shaped convex shape at the joint, the leg part is in a V-shaped concave shape at the joint, and the V-shaped convex shape is matched with the V-shaped concave shape, so that the joint of the main body part and the leg part is in a V shape; the friction coefficient of the V-shaped concave front inclined surface is smaller than that of the rear inclined surface, and the power assembly is an ejection assembly; the front inclined surface refers to an inclined surface closer to the front of the split type simulation robot in the two inclined surfaces in the V-shaped concave shape, and the rear inclined surface refers to an inclined surface closer to the rear of the split type simulation robot in the two inclined surfaces in the V-shaped concave shape;
the step S6 of controlling the power assemblies respectively disposed on the two legs to make the split type simulation robot obtain the upward first momentum and the forward second momentum includes:
and S601, controlling the ejection assemblies respectively arranged on the two legs, so that the ejection assemblies give upward first momentum and forward second momentum to the main body part of the split type simulation robot by using the V-shaped concave backward inclined surface and the V-shaped convex shape of the main body part as force applying surfaces and by using the V-shaped concave forward inclined surface as a track.
Thus, the stability of the main body part flying above the obstacle is ensured by the special V-shaped design. Because the main body part is in a V-shaped convex shape at the joint and the leg part is in a V-shaped concave shape at the joint, the friction coefficient of the V-shaped concave forward inclined surface is smaller than that of the backward inclined surface, the power assembly is an ejection assembly, so that the power assembly pushes the main body part from the V-shaped concave backward inclined surface and takes the V-shaped concave forward inclined surface as a guide track, therefore, the main body part obtains a comprehensive vector, the comprehensive vector can be decomposed into an upward first momentum and a forward second momentum, and the friction coefficient of the V-shaped concave forward inclined surface is smaller than that of the backward inclined surface, that is, the V-shaped recessed forward-inclined surface is smooth and does not cause a large amount of momentum loss, so that the value and direction of the resultant vector obtained by the main body portion can be accurately controlled, thereby enabling the main body portion to stably pass over an obstacle. It should be noted that, the V-shaped design of the present application, in addition to being beneficial to the precise control given by the momentum mentioned above so as to improve the stability of the main body in the air, is also beneficial to combining the main body with the two legs into the split type simulation robot in the subsequent steps (this is because, through the V-shaped design, when the main body descends, the requirement of the alignment accuracy of the main body and the legs is objectively reduced, only the horizontal position of the V-shaped convex shape of the main body is approximately the same as the horizontal position of the V-shaped concave shape of the legs, and then the main body can be conveniently combined with the two legs into the original split type simulation robot due to the objective characteristic of the V-shaped structure when the main body falls).
As described in the above steps S7-S9, sending an opening command to a second magnetic field generator preset under the floor of the first area to open the second magnetic field generator and to open the first magnetic field generator preset on the main body, so that the main body can be supported by magnetic force without falling onto an obstacle when passing over the first area; sending a rolling movement instruction to the two separated legs so that the two legs can bypass the obstacle in a rolling movement mode; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated; when the main body part crosses an obstacle from the air and the two leg parts move at the same speed in the horizontal direction under the main body part, the first magnetic field generator is closed, so that the main body part falls on the two leg parts to form the split type simulation robot, and a closing instruction is sent to the second magnetic field generator, so that the obstacle avoidance process of the split type simulation robot with the double walking modes is completed.
One difficulty with split obstacle avoidance is that the main body portion is prone to falling in the air when not passing over an obstacle. In order to prevent this, the present application adopts a special design that two magnetic field generators are provided to provide temporary magnetic buoyancy of the main body. Wherein the magnetic field generator may be any feasible device, for example, most simply, an electromagnetic induction coil or the like is used. Since the magnetic induction line generation technology and the magnetic induction line arrangement technology are the prior art, they are not described herein again. The main body portion can be supported magnetically when passing over the first area without falling onto an obstacle, which can be achieved in any way, for example, a first magnetic field generator on the main body portion generates a first magnetic flux line in a vertical direction (referring to the area where the obstacle is located) in a single case (i.e. without taking into account a second magnetic field generator), and correspondingly, a second magnetic field generator generates a second magnetic flux line in a vertical direction (also referring to a single case, i.e. without taking into account the first magnetic field generator), and controls the magnetic flux density (or magnetic field strength) so that the main body portion will be supported magnetically strong enough in a vertical direction. In addition, a rolling movement instruction is sent to the two separated legs, so that the two legs can bypass the obstacle in a rolling movement mode; wherein, hidden roller skate part has all been preset to every shank can roll the removal after the separation. The hidden roller-skating component of the leg can be hidden at the sole by adopting a telescopic structure, and when rolling is needed, the roller-skating component is extended out, so that rolling movement can be carried out (of course, the leg is preset with a motor and other devices for providing power). At the moment, the space required by the legs in the process of bypassing is far smaller than that of the simulation robot, so that the obstacle avoidance can be realized by only using a small space. In addition, the leap-type obstacle avoidance method is not adopted, and the reason is that the simulation robot is not a real person after all, and when the leap-type obstacle avoidance method is adopted, the posture of the simulation robot in the air is kept to be a big problem, so that the obstacle avoidance is difficult to ensure safely; in the split type obstacle avoidance method, only the main body part crosses the obstacle avoidance in the air, so that the posture of the main body part can be accurately controlled (the reason is described in the V-shaped structure, and although the effect of using the V-shaped structure is the best, the posture control effect of the split type obstacle avoidance method is also superior to that of leap type obstacle avoidance even though the V-shaped structure is not used). When the main body part crosses an obstacle from the air and the two leg parts move at the same speed in the horizontal direction under the main body part, the first magnetic field generator is turned off, so that the main body part falls onto the two leg parts to form the split type simulation robot. The mode of forming the split type simulation robot is just opposite to the mode of separating the split type simulation robot, namely if the split type simulation robot is separated by adopting the mode of opening the electromagnetic valve, the split type simulation robot is combined by adopting the mode of closing the electromagnetic valve at the moment. And sending a closing instruction to the second magnetic field generator so as to complete the obstacle avoidance process of the split type simulation robot with the double walking modes.
Further, a track network is pre-arranged under the floor, a vehicle carrying a second magnetic field generator moves along the track network, and the track network is made of a non-magnetic material, and the track network sends an opening command to the second magnetic field generator preset under the floor of the first area to open the second magnetic field generator and open the first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force when passing over the first area and cannot fall onto an obstacle before the step S7, the method includes:
s61, sending the position information of the first area to the vehicle, and requesting the vehicle to obtain the position information of the first area,
moving along the network of tracks to under the floor of the first area.
Thus, only one second magnetic field generator needs to be preset, and the magnetic field generator moves to the first area and provides a temporary magnetic field when the temporary magnetic field is needed in the first area. Thereby saving the cost of laying multiple magnetic field generators.
The obstacle avoidance method of the split type simulation robot with the double walking modes is realized by adopting the design of the split type simulation robot with the double walking modes, so that the obstacle avoidance operation can be carried out efficiently even when the narrow passage (or the cross-sectional area available for the bypassing at the obstacle side is smaller than that of the simulation robot) is in contact with an obstacle. Specifically, when the obstacle is touched and the obstacle is recorded in a preset split type obstacle avoidance list, the main body part of the split type simulation robot can cross over the obstacle, the legs of the split type simulation robot can cross over the obstacle in a rolling movement mode from the side of the obstacle, and the split type simulation robot is formed again after the obstacles are crossed, so that the obstacle avoidance method is high in adaptability due to the bypassing mode, and when the main body part of the split type simulation robot crosses over the obstacle, the main body part of the split type simulation robot is prevented from falling onto the obstacle in a temporary magnetic field supporting mode, and the overall safety is improved.
The embodiment of the application provides an obstacle avoidance device of a split type simulation robot with double walking modes, wherein the split type simulation robot is composed of two legs which are detachably connected and a main body part above the two legs, the two legs are respectively and detachably and mechanically connected with the main body part, and first magnetic field generators are arranged at the joints of the main body part and the two legs respectively; the two legs are provided with built-in chips, so that after the two legs are separated from the main body part respectively, the two legs can execute operation corresponding to the control signal according to the received control signal; the device comprises:
the system comprises a first environment signal sequence acquisition unit, a second environment signal sequence acquisition unit and a control unit, wherein the first environment signal sequence acquisition unit is used for enabling the split type simulation robot to move along a preset track in a preset stepping walking mode, and meanwhile, a first area in front of the split type simulation robot is subjected to real-time environment signal detection by utilizing a preset environment detection sensor array on the split type simulation robot, so that a first environment signal sequence is obtained;
the analysis result acquisition unit is used for inputting the first environment signal sequence into a preset obstacle analysis model for processing so as to obtain an analysis result output by the obstacle analysis model; the analysis result is that the obstacle exists or does not exist, the obstacle analysis model is based on a neural network model and is trained by adopting training data, and the training data is composed of a signal sequence for training and an artificial label corresponding to the signal sequence for training;
an analysis result judgment unit for judging whether the analysis result is an obstacle;
the obstacle classification unit is used for carrying out obstacle classification processing according to a preset obstacle classification method if the analysis result shows that the obstacle exists, so that a first obstacle class corresponding to the obstacle in the first area is obtained;
the split type obstacle avoidance list judging unit is used for judging whether the first obstacle category is recorded in a preset split type obstacle avoidance list or not;
the simulation robot separation unit is used for separating the two leg parts from the main body part and controlling power components respectively arranged on the two leg parts if the first obstacle category is recorded in a preset split type obstacle avoidance list so as to enable the split type simulation robot to obtain an upward first momentum and a forward second momentum;
the opening instruction sending unit is used for sending an opening instruction to a second magnetic field generator preset below the floor of the first area so as to open the second magnetic field generator and open a first magnetic field generator preset on the main body part, so that the main body part can be supported by magnetic force and cannot fall onto an obstacle when passing over the first area;
a rolling movement instruction sending unit, configured to send a rolling movement instruction to the two separated leg portions, so that the two leg portions bypass the obstacle in a rolling movement manner; each leg part is preset with a hidden roller sliding part, so that each leg part can roll and move after being separated;
and the simulation robot combination unit is used for closing the first magnetic field generator when the main body part crosses an obstacle from the air and the two legs move at the same speed in the horizontal direction under the main body part and with the main body part, so that the main body part falls on the two legs to form the split simulation robot, and sending a closing instruction to the second magnetic field generator, thereby completing the obstacle avoidance process of the split simulation robot with the double walking mode.
The operation performed by each of the units corresponds to the step of the obstacle avoidance method of the split type simulation robot with the dual walking mode in the foregoing embodiment one to one, and is not described herein again.
The utility model provides a keep away barrier device of split type emulation robot with two walking modes through the design that adopts the split type emulation robot of two walking modes to realize the method of keeping away the barrier of split type emulation robot, thereby even when meeting the barrier in narrow passageway department (or when the cross sectional area that can supply to detour in the obstacle side is less than the cross sectional area of emulation robot), also can carry out the high-efficient operation of keeping away the barrier. Specifically, when the obstacle is touched and the obstacle is recorded in a preset split type obstacle avoidance list, the main body part of the split type simulation robot can cross over the obstacle, the legs of the split type simulation robot can cross over the obstacle in a rolling movement mode from the side of the obstacle, and the split type simulation robot is formed again after the obstacles are crossed, so that the obstacle avoidance method is high in adaptability due to the bypassing mode, and when the main body part of the split type simulation robot crosses over the obstacle, the main body part of the split type simulation robot is prevented from falling onto the obstacle in a temporary magnetic field supporting mode, and the overall safety is improved.
Referring to fig. 2, an embodiment of the present invention further provides a computer device, where the computer device may be a server, and an internal structure of the computer device may be as shown in the figure. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer equipment is used for storing data used by the obstacle avoidance method of the split type simulation robot with the double walking mode. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to realize an obstacle avoidance method of the split type simulation robot with the double walking modes.
The processor executes the obstacle avoidance method of the split type simulation robot with the double walking modes, wherein the steps included in the method correspond to the steps of executing the obstacle avoidance method of the split type simulation robot with the double walking modes in the embodiment one to one, and the details are not repeated herein.
It will be understood by those skilled in the art that the structures shown in the drawings are only block diagrams of some of the structures associated with the embodiments of the present application and do not constitute a limitation on the computer apparatus to which the embodiments of the present application may be applied.
The computer equipment realizes the obstacle avoidance method of the split type simulation robot by adopting the design of the split type simulation robot with the double walking modes, so that the obstacle avoidance operation can be carried out efficiently even when the narrow passage (or the cross section area available for bypassing at the obstacle side is smaller than that of the simulation robot) touches the obstacle. Specifically, when the obstacle is touched and the obstacle is recorded in a preset split type obstacle avoidance list, the main body part of the split type simulation robot can cross over the obstacle, the legs of the split type simulation robot can cross over the obstacle in a rolling movement mode from the side of the obstacle, and the split type simulation robot is formed again after the obstacles are crossed, so that the obstacle avoidance method is high in adaptability due to the bypassing mode, and when the main body part of the split type simulation robot crosses over the obstacle, the main body part of the split type simulation robot is prevented from falling onto the obstacle in a temporary magnetic field supporting mode, and the overall safety is improved.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored thereon, and when the computer program is executed by a processor, the method for avoiding an obstacle of the split-type simulation robot with the dual-walking mode is implemented, where steps included in the method are respectively one-to-one corresponding to steps of the method for avoiding an obstacle of the split-type simulation robot with the dual-walking mode in the foregoing embodiment, and are not described herein again.
The computer-readable storage medium of the application realizes the obstacle avoidance method of the split type simulation robot by adopting the design of the split type simulation robot with the double walking mode, so that the obstacle avoidance operation can be carried out efficiently even if the narrow passage (or the cross-sectional area available for bypassing at the obstacle side is smaller than that of the simulation robot) meets the obstacle. Specifically, when the obstacle is touched and the obstacle is recorded in a preset split type obstacle avoidance list, the main body part of the split type simulation robot can cross over the obstacle, the legs of the split type simulation robot can cross over the obstacle in a rolling movement mode from the side of the obstacle, and the split type simulation robot is formed again after the obstacles are crossed, so that the obstacle avoidance method is high in adaptability due to the bypassing mode, and when the main body part of the split type simulation robot crosses over the obstacle, the main body part of the split type simulation robot is prevented from falling onto the obstacle in a temporary magnetic field supporting mode, and the overall safety is improved.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware associated with a computer program or instructions, the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program can include the processes of the embodiments of the methods described above when executed. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.