Method for determining collision position and robot

Abstract

The invention relates to the technical field of robots, and discloses a collision position determination method and a robot. The robot comprises a collision part and at least two collision sensors, the at least two collision sensors are installed on the collision part at intervals, and the method comprises the steps that a collision signal is obtained and collected by the collision sensors, and collision evaluation information of the collision sensors is determined according to the collision signal, and determining a target collision area according to the collision evaluation information of each collision sensor, the target collision area being an area meeting a first maximum likelihood condition of collision in the collision part, and determining a collision position according to the target collision area. According to the embodiment of the invention, the collision evaluation information of each collision sensor can be synthesized to determine the collision position, and the collision position output by the mode is not limited to the area with the collision sensor or can be the area without the collision sensor, so that the collision position can be accurately and reliably determined, and the collision accuracy is improved. And the fine granularity for determining the collision position can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of robots, and particularly relates to a method for determining a collision position and a robot. Background Art

[0002] Robots provided by related technologies are provided with a plurality of collision sensors on a collision part. When the collision part of the robot collides with an obstacle, the collision sensors collect collision signals, and the robot determines the collision position according to the collision signals.

[0003] The robots provided by related technologies determine the collision position in a plurality of installation areas corresponding to the plurality of collision sensors. When a collision occurs in the middle of the installation areas corresponding to two adjacent collision sensors, related technologies determine the collision position between the installation areas corresponding to two adjacent collision sensors according to the maximum signal energy principle. For example, if a collision occurs in the middle of the installation area corresponding to collision sensor G1 and the installation area corresponding to collision sensor G2, and the signal energy detected by collision sensor G1 is greater than the signal energy detected by collision sensor G2, related technologies determine the installation area corresponding to collision sensor G1 as the collision position. However, in fact, the collision occurs in the middle of the installation area corresponding to collision sensor G1 and the installation area corresponding to collision sensor G2. Therefore, related technologies cannot accurately and reliably determine the collision position. Summary of the Invention

[0004] An object of an embodiment of the present application is to provide a method for determining a collision position and a robot, so as to solve the technical problem that related technologies cannot accurately and reliably determine the collision position.

[0005] In a first aspect, an embodiment of the present application provides a method for determining a collision position, which is applied to an electronic device. The electronic device includes a collision part and at least two collision sensors, and the at least two collision sensors are installed on the collision part at intervals, including:

[0006] Obtain a collision signal, where the collision signal is collected by the collision sensor;

[0007] Determine collision evaluation information of the collision sensor according to the collision signal;

[0008] Determine a target collision area according to the collision evaluation information of each collision sensor, where the target collision area is an area in the collision part that satisfies a first maximum likelihood condition for a collision to occur;

[0009] Determine the collision position according to the target collision area.

[0010] Optionally, the determining the collision evaluation information of the collision sensor according to the collision signal includes:

[0011] Determine the collision signal with a frequency in the target frequency band according to the collision signal as the target signal;

[0012] Collect a plurality of data points on the target signal according to a first preset window length;

[0013] Determine collision evaluation information corresponding to the first preset window length according to the plurality of data points of the collision sensor.

[0014] Optionally, the target signal is a quasi-static force signal or a static force signal, and the target frequency band is [0Hz, 150Hz].

[0015] Optionally, each collision sensor is configured with a detection partition on the collision part, and determining the target collision area according to the collision evaluation information of each collision sensor includes:

[0016] Determine at least one detection partition that satisfies the first maximum likelihood condition among the plurality of detection partitions as the target detection partition according to the collision evaluation information of each collision sensor, where the target detection partition is the target collision area.

[0017] Optionally, the collision evaluation information includes a first type of collision evaluation value, and determining at least one detection partition that satisfies the first maximum likelihood condition among the plurality of detection partitions as the target detection partition according to the collision evaluation information of each collision sensor includes:

[0018] Select two first type of collision evaluation values with the largest numerical values from among the first type of collision evaluation values of each to obtain a first evaluation value and a second evaluation value, where the first evaluation value is greater than the second evaluation value;

[0019] Determine whether the detection partition corresponding to the first evaluation value is adjacent to the detection partition corresponding to the second evaluation value;

[0020] If they are adjacent, determine that both the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value satisfy the first maximum likelihood condition and are both target detection partitions.

[0021] Optionally, the collision evaluation information further includes a second type of collision evaluation value. If the detection partition corresponding to the first evaluation value is not adjacent to the detection partition corresponding to the second evaluation value, before determining the collision position according to the target detection partition, the method further includes:

[0022] Obtain the second type of collision evaluation values of each collision sensor;

[0023] Select the two largest second type of collision evaluation values from among the second type of collision evaluation values of each to obtain a third evaluation value and a fourth evaluation value, where the third evaluation value is greater than the fourth evaluation value;

[0024] Determine whether the detection partition corresponding to the third evaluation value is adjacent to the detection partition corresponding to the fourth evaluation value;

[0025] If they are adjacent, determine that both the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value satisfy the first maximum likelihood condition and are both target detection partitions.

[0026] Optionally, the method further includes:

[0027] If the detection partition corresponding to the third evaluation value is not adjacent to the detection partition corresponding to the fourth evaluation value, determine that the detection partition corresponding to the first evaluation value is the target detection partition.

[0028] Optionally, the detection partition includes a first detection sub-region and at least one second detection sub-region, the second detection sub-region is adjacent to the first detection sub-region, the collision sensor is disposed on the first detection sub-region, and the determining the collision position according to the target collision region includes:

[0029] Determine, in at least one of the target detection partitions, a detection sub-region that satisfies the second maximum likelihood condition for a collision to occur as a reference detection sub-region;

[0030] Determine the collision position according to the position of the reference detection sub-region in the collision part.

[0031] Optionally, the determining, in at least one of the target detection partitions, a detection sub-region that satisfies the second maximum likelihood condition for a collision to occur as a reference detection sub-region includes:

[0032] If the number of the target detection partitions is at least two, calculate the ratio of the first type of collision evaluation value of the first reference partition to the first type of collision evaluation value of the second reference partition. The first reference partition is the target detection partition with the maximum first type of collision evaluation value among at least one of the target detection partitions, and the second reference partition is the target detection partition in at least one of the target detection partitions where the first type of collision evaluation value is smaller than the first type of collision evaluation value of the first reference partition but larger than the first type of collision evaluation values of the remaining target detection partitions. Or, the first reference partition is the target detection partition with the maximum second type of collision evaluation value among at least one of the target detection partitions, and the second reference partition is the target detection partition in at least one of the target detection partitions where the second type of collision evaluation value is smaller than the second type of collision evaluation value of the first reference partition but larger than the second type of collision evaluation values of the remaining target detection partitions;

[0033] If the ratio is less than a preset ratio threshold, it is determined that the second detection sub-region adjacent to the second reference partition in the first reference partition satisfies the second maximum likelihood condition of a collision, and the second detection sub-region is a reference detection sub-region;

[0034] If the ratio is not less than the preset ratio threshold, it is determined that the first detection sub-region of the first reference partition satisfies the second maximum likelihood condition of a collision, and the first detection sub-region is a reference detection sub-region.

[0035] Optionally, the determining the detection sub-region that satisfies the second maximum likelihood condition of a collision in at least one of the target detection partitions as the reference detection sub-region includes:

[0036] If the number of the target detection partitions is 1, it is determined that the first detection sub-region of the target detection partition satisfies the second maximum likelihood condition of a collision, and the first detection sub-region of the target detection partition is a reference detection sub-region.

[0037] Optionally, the first type of collision evaluation value is a signal energy value, and the second type of collision evaluation value is a signal peak value.

[0038] Optionally, the collision evaluation information includes a first type of collision evaluation value and a second type of collision evaluation value. Before determining the target collision region, the method further includes:

[0039] Calculating the sum of the first type of collision evaluation values of each of the collision sensors to obtain a total collision evaluation value;

[0040] Selecting the maximum second type of collision evaluation value from the second type of collision evaluation values of each of the collision sensors;

[0041] Generating collision effect information according to the maximum second type of collision evaluation value and the total collision evaluation value;

[0042] Controlling whether the electronic device continues to perform the collision detection operation according to the collision effect information.

[0043] Optionally, the collision effect information includes collision valid information and collision invalid information. The generating the collision effect information according to the maximum second type of collision evaluation value and the total collision evaluation value includes:

[0044] Judging whether the total collision evaluation value is greater than a first preset evaluation threshold and whether the maximum second type of collision evaluation value is greater than a second preset evaluation threshold;

[0045] If both are greater, it is determined that the collision effect information is collision valid information;

[0046] If not, it is determined that the collision effect information is collision invalid information.

[0047] Optionally, the frequency of the target signal is in the range of [0Hz, 150Hz].

[0048] In a second aspect, an embodiment of the present application provides a robot, including:

[0049] A collision part;

[0050] A plurality of collision sensors, which are installed at intervals on the collision part;

[0051] A signal conditioning circuit, which is electrically connected to each of the collision sensors;

[0052] A controller, which is electrically connected to the signal conditioning circuit and is used to execute the above-mentioned method for determining the collision position.

[0053] In a third aspect, an embodiment of the present application provides a non-volatile readable storage medium, and the non-volatile readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the above-mentioned method for determining the collision position.

[0054] In the method for determining the collision position provided by the embodiment of the present application, a collision signal is obtained. Among them, the robot includes a collision part and at least two collision sensors, and the at least two collision sensors are installed at intervals on the collision part. The collision signal is collected by the collision sensors. The collision evaluation information detected by the collision sensors is determined according to the collision signal, and the target collision area is determined according to the collision evaluation information of each collision sensor. The target collision area is the area in the collision part that satisfies the first maximum likelihood condition for a collision to occur. The collision position is determined according to the target collision area. This embodiment can comprehensively determine the collision position based on the collision evaluation information of each collision sensor. In this way, the output collision position is not limited to the area where the collision sensor is provided, but can also be the area where there is no collision sensor. In this way, not only can the collision position be accurately and reliably determined, but also the granularity of determining the collision position can be improved. Description of the Drawings

[0055] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings. These exemplary illustrations do not limit the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements, unless otherwise stated, the drawings in the drawings do not constitute a proportional limitation.

[0056] Figure 1 It is a schematic structural diagram of the robot provided by the embodiment of the present application;

[0057] Figure 2 It is a circuit principle block diagram of a robot provided by the embodiment of the present application;

[0058] Figure 3A frequency response diagram when a robot collides with hard and soft obstacles provided by an embodiment of the present application;

[0059] Figure 4 A schematic flowchart of a method for determining a collision position provided by an embodiment of the present application;

[0060] Figure 5a A schematic diagram of a low-pass frequency response curve of a FIR filter provided by an embodiment of the present application;

[0061] Figure 5b A schematic structural diagram of a collision part provided by an embodiment of the present application;

[0062] Figure 6 For Figure 5b The effect diagram of unfolding the collision part of into a plane;

[0063] Figure 7 Based on Figure 6 The schematic diagram of further subdividing the detection sub-areas for the provided detection partition;

[0064] Figure 8 A schematic flowchart for determining the collision position provided by an embodiment of the present application;

[0065] Figure 9 A schematic diagram of a target signal of a complete collision duration provided by an embodiment of the present application;

[0066] Figure 10 A schematic diagram of collecting data by using a second preset window length of L in the related art;

[0067] Figure 11 A schematic diagram of collecting data by using a second preset window length of 2L provided by an embodiment of the present application;

[0068] Figure 12 A schematic structural diagram of a controller provided by an embodiment of the present application. Detailed implementation manners

[0069] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0070] It should be noted that if there is no conflict, the various features in the embodiments of the present application can be combined with each other, and all are within the protection scope of the present application. In addition, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flowchart. Furthermore, the terms "first", "second", "third", etc. used in the present application do not limit the data and the execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0071] In addition to discovering the technical problems pointed out in the background art, the applicant also found during the research on the collision detection scheme that:

[0072] The mechanical collision sensors provided by the related art can detect collisions using the infrared detection principle or based on the displacement stroke principle. The mechanical collision sensor based on the infrared detection principle is installed inside the intelligent mobile robot. Among them, the mechanical collision sensor includes an infrared emitter and an infrared receiver. The infrared emitter and the infrared receiver are arranged on the inner side of the front windshield. A collision plate is provided between the infrared emitter and the infrared receiver. When the front windshield is collided by an obstacle, the front windshield drives the collision plate to block the light path between the infrared emitter and the infrared receiver, and the infrared receiver fails to receive the signal emitted by the infrared emitter. Therefore, the controller detects whether the front windshield collides with an obstacle according to the signal received by the infrared emitter.

[0073] Generally, a plurality of mechanical collision sensors are arranged on the front windshield of the sweeping robot to detect the collision position. However, when the front windshield is collided by an obstacle, the front windshield will generate displacement, which will cause a plurality of mechanical collision sensors at different positions to generate collision signals, and the collision signals are likely to reduce the accuracy of collision detection.

[0074] The mechanical collision sensor based on the displacement stroke principle is installed in the split front windshield structure. Specifically, the self-moving device is provided with a split front windshield structure, and the split front windshield structure includes a front side plate and a rear side plate. The mechanical collision sensor is usually installed between the front side plate and the rear side plate. When the front side plate collides with an obstacle, the obstacle pushes the front side plate to move in the direction of the rear side plate. Then, the front side plate pushes the mechanical collision sensor to move along, so that the mechanical collision sensor generates a displacement stroke to further output a collision signal.

[0075] As can be seen from the above description, in order to meet the requirement that the mechanical collision sensor needs to have a displacement stroke, the related technology usually adopts a split front baffle structure. This will result in a relatively large overall housing of the intelligent mobile robot, and it is necessary to manually assemble the mechanical collision sensor between the front side plate and the rear side plate. The manual assembly accuracy is different, resulting in unstable detection performance of the mechanical collision sensor, and it has a high requirement for the proficiency of manual assembly, with low assembly efficiency and high cost.

[0076] In addition, when the robot performs operations such as crossing a threshold or other operations during the working process, the chassis of the robot is prone to collide with the ground, and this collision is also likely to cause signal crosstalk. Moreover, the high-frequency vibration caused by the motor vibration and sudden start and stop of the robot is also likely to cause signal crosstalk, resulting in the robot being unable to reliably and accurately determine the collision position.

[0077] The applicant also found another related technology. In another related technology, a flexible piezoelectric material is pasted all around the front baffle. The flexible piezoelectric material is used to detect the change in pressure on the front baffle to determine whether a collision occurs and the collision position. The cost of such a technical solution is relatively high, and there are relatively high requirements for the pasting position of the flexible piezoelectric material and the appearance design. At the same time, when crossing a threshold during the traveling process, it may cause the flexible piezoelectric material to generate a pressure signal and cause false alarms.

[0078] The applicant also found another related technology. In another related technology, a plurality of pressure sensors are arranged on the front baffle or the body of the robot. When a collision occurs, the pressure sensors output collision signals for detecting the collision. This solution cannot effectively identify light collision scenarios or slow collision scenarios.

[0079] The applicant also found another related technology. In another related technology, a mechanical collision sensor is arranged on the front baffle of the robot. The robot detects whether it encounters an obstacle according to the collision signal collected by the mechanical collision sensor. When the robot collides with an obstacle made of a hard material, the robot can easily determine whether a collision with the obstacle occurs according to such a collision signal. However, when the robot collides with an obstacle made of a soft material, the robot is not easy to identify the collision signal at this time, resulting in the robot being unable to detect whether a collision with the obstacle has occurred, and further resulting in the robot being unable to take obstacle avoidance measures in time.

[0080] In addition, the applicant also found that: in the related art, it is necessary to collect a complete collision signal for collision detection. Usually, the collision duration of a complete collision signal is greater than 200 ms. Adding the time for operations such as sliding window acquisition, the time required for collision detection based on a complete collision signal is usually greater than 400 ms. Usually, the minimum moving speed of a robot indoors is about 0.1 m / s. The moving distance corresponding to a duration of 400 ms is 4 cm. A distance of 4 cm is relatively long for a robot. Before the collision avoidance action arrives after the robot collides with an object, the robot can push the object and move 4 cm until the object has a rigid collision with the wall. Thus, it is easy to damage the object. Therefore, the collision detection time in the related art is relatively long, resulting in a low collision avoidance performance of the robot.

[0081] In the embodiment of the present application, by using a quasi-static force signal or a static force signal as a target signal to judge a collision, collision detection can be simply realized without a complex structural design. In addition, when the robot collides with an obstacle of any material, a quasi-static force signal or a static force signal will be generated. That is, the quasi-static force signal or the static force signal can reflect the collision between the robot and an obstacle of any material. Therefore, in the embodiment of the present application, collision is detected according to the target signal, so that collision detection between the self-mobile device and obstacles of all materials can be realized, which is conducive to the self-mobile device taking corresponding collision avoidance measures reliably and effectively.

[0082] In addition, there is a regular linear relationship between the amplitude of the target signal corresponding to collision sensors at different positions and the collision position. For example, the amplitude of the target signal is negatively correlated with the relative distance of the collision position. The amplitude of the target signal corresponding to the collision sensor closer to the collision position is larger, and the amplitude of the target signal corresponding to the collision sensor farther from the collision position is smaller. Based on the target signal that changes according to this rule, in this embodiment, the collision position can be reliably and accurately judged, effectively solving the technical problem of signal crosstalk.

[0083] At the same time, as mentioned above, the target signal can not only reliably detect a collision in both a light collision scenario or a slow collision scenario, but also in this embodiment, data points of the target signal can be collected with a window length smaller than the collision duration of the target signal. Thus, not only can a collision be reliably detected, but also the collision detection time can be shortened, improving the collision detection efficiency.

[0084] In addition, in the embodiment of the present application, the area where the collision part can collide with an obstacle is divided into multiple large detection partitions, and each large detection partition is further subdivided into multiple small detection partitions. Thus, the collision position can be determined more precisely and with finer granularity, avoiding the situation in the related art where only a rough collision area can be provided and the precise collision position cannot be improved.

[0085] In addition, in this embodiment, instead of using the target signal of the complete collision duration for collision detection, a short window method is adopted for collision detection. This embodiment can perform collision detection faster, shorten the collision response time, and has a relatively high detection efficiency. In addition, this embodiment uses the trigger edge detection method for collision detection. Compared with the related technology that needs to calculate the sum of the energies of the entire window, this embodiment has a high calculation efficiency, which is beneficial to improving the speed and efficiency of collision detection.

[0086] An embodiment of the present application provides a robot, which can be a cleaning robot, a handling robot, a guiding robot, etc. The cleaning robot includes a sweeping robot, a mopping robot, a sweeping and mopping integrated robot, a floor washing robot, etc.

[0087] Please refer to Figure 1 and Figure 2 , the robot 100 includes a collision part 11, a plurality of collision sensors 12, a signal conditioning circuit 13 and a controller 14.

[0088] The collision part 11 can be a front bumper, a front collision, a front impact or other structures for buffering the collision of the robot. Among them, the collision part 11 includes a side shell 111, an upper shell 112 and a connecting member 113. The side shell 111 is detachably installed on the outer periphery of the robot 100. Among them, the side shell 111 can surround the outer periphery of the robot 100. When an obstacle collides with the robot 100, the side shell 111 contacts the obstacle first as a buffer plate, so that the internal components of the robot 100 can be protected from being damaged.

[0089] The upper shell 112 is used to be installed on the top of the robot 100. Among them, the upper shell 112 is connected to the side shell 111.

[0090] The connecting member 113 is connected between the side shell 111 and the upper shell 112, and the collision sensor 12 is arranged on the connecting member 113. As Figure 1 shown, there is a gap 114 between the side shell 111 and the upper shell 112. One end of the connecting member 113 is connected to the upper shell 112, and the other end of the connecting member 113 extends away from the upper shell 112 and passes through the gap 114 to be connected to the side shell 111.

[0091] As described above, since there is a gap 114 between the upper housing 112 and the side housing 111, and the connecting member 113 is equivalent to bridging across the gap 114 to connect the side housing 111 and the upper housing 112, the connecting member 113 can be equivalent to a bridge arm between the side housing 111 and the upper housing 112. When the side housing 111 collides with an obstacle, since the gap 114 can reduce the rigidity of the side housing 111, when the side housing 111 is squeezed by the obstacle, the stress will be quickly and comprehensively concentrated on the connecting member 113, so that the connecting member 113 absorbs the kinetic energy when the side housing 111 is collided, and can respond to the stress with high sensitivity and cause its own deformation stress. The collision sensor 12 provided on the connecting member 113 detects a large deformation stress and generates a collision signal.

[0092] A plurality of collision sensors 12 are installed at intervals on the collision part 11. Among them, the collision sensors 12 can be integrally injection-molded with the collision part 11, or a plurality of collision sensors 12 can be arranged at intervals along the circumference of the collision part 11 by an external mounting method.

[0093] It can be understood that the collision sensor 12 includes at least one of a piezoelectric ceramic sensor, a piezoelectric ceramic sensor, a relaxor ferroelectric sensor, a bone voiceprint sensor, an acceleration sensor, a gyroscope, and a strain gauge.

[0094] It can also be understood that the collision sensor 12 can be designed into a detection structure that realizes the detection function of one collision position, or the collision sensor 12 can also be designed into a detection structure that realizes the detection function of two or more collision positions. This detection structure is an integrated appearance structure, and this detection structure can detect collisions from different positions, that is, this detection structure can be configured with multiple input ends and a single output end or multiple input ends and multiple output ends.

[0095] The signal conditioning circuit 13 is electrically connected to each collision sensor 12 and is used for signal conditioning of the collision signal to obtain a conditioned collision signal.

[0096] Please continue to refer to Figure 2 , the signal conditioning circuit 13 includes a charge amplification circuit 131 and a low-pass filter circuit 132. The charge amplification circuit 131 is electrically connected to the collision sensor 12 and is used for amplifying the collision signal. The low-pass filter circuit 132 is electrically connected to the charge amplification circuit 131 and the controller 14 respectively and is used for filtering the amplified collision signal.

[0097] The controller 14 is electrically connected to the signal conditioning circuit 13 and is configured to process the conditioned collision signal to obtain a target signal with a frequency within a target frequency band, where the target signal can be a quasi-static force signal or a static force signal. A static force signal is a force signal whose amplitude basically does not change with time. For example, when the robot collides with an obstacle and the collision surface where the robot collides with the obstacle is in close contact with the obstacle, the deformation stress of the collision part 11 remains unchanged. Therefore, the collision sensor 12 collects the collision signal, and the controller 14 can obtain the static force signal from the collision signal. Among them, the amplitude of the static force signal does not change with time. Therefore, the frequency of the static force signal is 0 Hz.

[0098] Quasi-static is a motion state between static and dynamic. A quasi-static force signal is a signal between a static force signal and a dynamic force signal. A quasi-static force signal refers to a signal with a frequency in the frequency range of (0 Hz, N Hz]. It can be understood that regardless of whether the robot collides with an obstacle of any material at a high speed or a low speed, the robot can obtain a quasi-static force signal with a frequency in the frequency range of (0 Hz, N Hz] from the collision signal corresponding to the high speed, or can obtain a quasi-static force signal with a frequency in the frequency range of (0 Hz, N Hz] from the collision signal corresponding to the low speed.

[0099] In some embodiments, the frequency of the quasi-static force signal is in the frequency range of (0 Hz, 150 Hz]. It can be seen that in this embodiment, N takes the value of 150. For the upper limit value of this frequency range, in different application scenarios and / or for different robots, the upper limit value of this frequency range can be customized by the designer according to engineering experience, or the upper limit value of this frequency range can also be selected according to the results of multiple test trials. For example, the upper limit value of this frequency range can be selected as different values, such as 5, 10, 20, 50, 70, 90, 130.

[0100] Please refer to Figure 3 , when the robot collides with obstacles of different materials, the frequency and amplitude of the collision signal are related to the softness and hardness of the obstacle material. As Figure 3 shown, when the robot collides with a hard obstacle, both the frequency and amplitude of the collision signal are relatively high. Among them, the first curve 31 is used to represent the change of the amplitude of the collision signal with time during a hard collision. The first curve 31 includes a first transient collision area 311 and a first quasi-static force area 312. The frequency of the collision signal in the first transient collision area 111 is relatively high, and the collision signal in the first quasi-static force area 312 is a quasi-static force signal.

[0101] When the robot collides with a soft obstacle, compared with the collision signal in the case of a hard obstacle, the frequency and amplitude of the collision signal corresponding to the soft case are relatively low. Among them, the second curve 32 is used to represent the change of the amplitude of the collision signal during a soft collision over time. The second curve 32 includes a second transient collision region 321 and a second quasi-static force region 322. The frequency of the collision signal in the second transient collision region 321 is relatively high, and the collision signal in the second quasi-static force region 322 is a quasi-static force signal. Among them, the frequency of the collision signal in the first transient collision region 311 is higher than that of the collision signal in the second transient collision region 321. However, the frequencies and amplitudes of the quasi-static force signals in the first quasi-static force region 312 and the second quasi-static force region 322 are relatively close. This shows that the static force signal or quasi-static force signal is independent of the material of the obstacle.

[0102] Since the static force signal or quasi-static force signal is independent of the material of the obstacle, whether the material of the obstacle is a hard material or a soft material, the static force signal or quasi-static force signal can characterize the collision between the robot and the obstacle, avoiding the situation where it is easy to detect no collision when the robot collides with an obstacle made of a soft material, which is beneficial to improving the reliability of collision detection and reducing the false detection rate.

[0103] When the robot collides with an obstacle, although the vibration from the collision position can be transmitted to adjacent collision sensors, resulting in signal crosstalk, the static force signal or quasi-static force signal (target signal) is not affected by this. As mentioned above, the amplitude of the target signal is negatively correlated with the relative distance of the collision position, that is, the larger the relative distance, the smaller the amplitude of the static force signal or quasi-static force signal, and the smaller the relative distance, the larger the amplitude of the static force signal or quasi-static force signal.

[0104] For example, three collision sensors are arranged at intervals along the circumference of the housing assembly 400, namely collision sensor A1, collision sensor A2, and collision sensor A3. Among them, the relative distance from the collision position to the position of collision sensor A1 on the housing assembly 400 is d1, the relative distance from the collision position to the position of collision sensor A2 on the housing assembly 400 is d2, and the relative distance from the collision position to the position of collision sensor A3 on the housing assembly 400 is d3, where d1 < d2 < d3.

[0105] The controller obtains the target signal k1 based on the collision signal collected by the collision sensor A1, obtains the target signal k2 based on the collision signal collected by the collision sensor A2, and obtains the target signal k3 based on the collision signal collected by the collision sensor A3. The amplitude of the target signal k1 is greater than the amplitude of the target signal k2, and the amplitude of the target signal k2 is greater than the amplitude of the target signal k3. Since the amplitude of the static force signal or quasi-static force signal is negatively correlated with the relative distance of the collision position, in this embodiment, the static force signal or quasi-static force signal that changes according to this law can be used to reliably and accurately determine the collision position even in the case of collision crosstalk, enhancing the ability to resist signal crosstalk between adjacent collision sensors.

[0106] As another aspect of the embodiments of the present application, the embodiments of the present application provide a method for determining a collision position, which is applied to an electronic device. It can be understood that the electronic device can be a robot, an independent collision component, or other devices. Please refer to Figure 4 , the method for determining the collision position includes the following steps:

[0107] S41: Obtain the collision signal.

[0108] In this step, the electronic device includes a collision part and at least two collision sensors. The at least two collision sensors are installed at intervals on the collision part. The collision signal is collected by the collision sensors. The electronic device polls each collision sensor according to a preset frequency to obtain the collision signals collected by each collision sensor.

[0109] S42: Determine the collision evaluation information of the collision sensor according to the collision signal.

[0110] In this step, the collision evaluation information is used to characterize whether the collision sensor detects a collision. In some embodiments, the collision evaluation information can be a type of collision evaluation value, and the collision evaluation value can be the signal energy, signal peak, or zero-crossing rate of the target signal, etc.

[0111] The level of the collision evaluation value can evaluate the degree of collision occurring in the installation area of the collision sensor. In this embodiment, a type of collision evaluation value is used as the collision evaluation information, which can effectively determine the collision position.

[0112] In some embodiments, the collision evaluation information includes at least two types of collision evaluation values, and the two types of collision evaluation values are any two of the evaluation values such as the signal energy, signal peak, or zero-crossing rate of the target signal. This embodiment can comprehensively consider multiple types of collision evaluation values and reliably determine the collision position.

[0113] S43: Determine the target collision area according to the collision evaluation information of each collision sensor. The target collision area is the area in the collision part that satisfies the first maximum likelihood condition for a collision to occur.

[0114] In this step, the first maximum likelihood condition is used to screen out the region corresponding to the maximum probability of collision in the collision part. Those skilled in the art can reasonably set the first maximum likelihood condition according to the content disclosed in this embodiment.

[0115] In some embodiments, the first maximum likelihood condition is that the collision evaluation value of the target collision sensor is the largest among the collision evaluation values of all collision sensors. Herein, the target collision sensor is one of at least two collision sensors. The region satisfying the first maximum likelihood condition of collision is the region where the target collision sensor is installed on the collision part.

[0116] In some embodiments, the first maximum likelihood condition is that the collision evaluation value of the first collision sensor is the largest among the collision evaluation values of all collision sensors, the collision evaluation value of the second collision sensor is smaller than that of the first collision sensor and larger than that of other collision sensors, the first collision sensor or the second collision sensor is one of at least two collision sensors, and the installation position of the first collision sensor is adjacent to that of the second collision sensor. The region satisfying the first maximum likelihood condition of collision is the region where the first collision sensor is installed on the collision part and the region where the second collision sensor is installed on the collision part.

[0117] S44: Determine the collision position according to the target collision region.

[0118] In this step, the collision position is used to represent the position where the collision occurs on the electronic device. In some embodiments, the collision position includes information such as right front, right rear, right of due front, or center of due front, etc. In some embodiments, the electronic device is configured with a reference line, and the collision position is the position corresponding to the angle deviating from the reference line. For example, if the angle of the reference line is 0 degrees, the collision position is the position corresponding to a 30-degree deviation from the reference line.

[0119] This embodiment can determine the collision position by comprehensively considering the collision evaluation information of each collision sensor. The collision position output in this way is not limited to the region where the collision sensor is provided, but can also be the region where no collision sensor is provided. In this way, not only can the collision position be accurately and reliably determined, but also the granularity of determining the collision position can be improved.

[0120] In some embodiments, determining the collision evaluation information of the collision sensor according to the collision signal includes the following steps:

[0121] S421: Determine that the collision signal with a frequency in the target frequency band is the target signal according to the collision signal.

[0122] S422: Collect a plurality of data points on the target signal according to the first preset window length;

[0123] S423: Determine the collision evaluation information corresponding to a first preset window length based on multiple data points of a collision sensor.

[0124] In S421, the target frequency band can be a high-frequency band or a low-frequency band. Preferably, in this embodiment, the low-frequency band can be selected as the target frequency band, and the target frequency band can be [0Hz, 150Hz]. Correspondingly, the target signal can be a high-frequency signal or a low-frequency signal. Preferably, in this embodiment, the low-frequency signal can be selected as the target signal, where the target signal is a quasi-static force signal or a static force signal.

[0125] It can be understood that, as described above, the amplitude of the target signal is negatively correlated with the relative distance of the collision position. In this embodiment, the target signal that changes according to this law can be used to reliably and accurately determine the collision position even in the case of signal crosstalk.

[0126] Determining that the collision signal with a frequency in the target frequency band is the target signal based on the collision signal includes: performing signal conditioning processing on the collision signal to obtain a conditioned collision signal, and performing digital low-pass filtering processing on the conditioned collision signal according to a preset digital low-pass filter to obtain a collision signal with a frequency in the target frequency band, and the collision signal with a frequency in the target frequency band is the target signal.

[0127] The preset digital low-pass filter can be an IIR filter (Infinite Impulse Response Filter) or an FIR filter (Finite Impulse Response Filter). The calculation complexity of the IIR filter is low. Please refer to Figure 5a , and the low-pass frequency response curve of the FIR filter is as Figure 5a shown, and the FIR filter has good stability and phase linearity.

[0128] The signal conditioning processing includes signal amplification processing and analog low-pass filtering processing. In this embodiment, the collision signal has been subjected to a low-pass filtering operation through an analog low-pass filter circuit. However, considering the influence of factors such as the filter order and filtering delay, the collision signals outside the specified low-frequency band still affect the subsequent collision detection and need to be filtered to reduce the influence of this part of the collision signals. In addition, the fan noise, traveling noise, and overstep noise of the sweeper are likely to interfere with the extraction of the quasi-static force signal. Therefore, in this embodiment, by setting the passband and stopband of the digital low-pass filter, such interference signals can be further filtered. In addition, if the digital low-pass filter is not used and only the analog low-pass filter circuit is used for low-pass filtering to filter various interference signals, on the one hand, the delay of this method is relatively large, and on the other hand, the device cost of this method increases. By compatible with the use of the digital low-pass filter in this embodiment, not only the influence of interference is effectively reduced, but also the device cost is reduced.

[0129] In S422, in this embodiment, sliding is performed on the target signal according to the first preset window length to obtain a plurality of data points. In some embodiments, the first preset window length is equal to the acquisition duration of the plurality of data points, and the acquisition duration is equal to the number of data points multiplied by the reciprocal of the acquisition frequency. For example, in this embodiment, 3 data points are acquired on the target signal according to the first preset window length, the acquisition frequency of each data point is 120 Hz, and the time for acquiring 3 data points is 3 / 120 = 25 ms. Therefore, the first preset window length provided by this embodiment can be 25 ms.

[0130] In some embodiments, the first preset window length is equal to the sum of the acquisition duration of the plurality of data points and a preset time margin. For example, the preset time margin is 25 ms, and the acquisition duration in the above example is 25 ms. Therefore, the first preset window length is 50 ms.

[0131] In some embodiments, the first preset window length is equal to the sum of the acquisition duration of the plurality of data points, the preset time margin, and the preset anti-interference time. For example, the preset anti-interference time is 50 ms, and this preset anti-interference time is the time for filtering the noise signal. Therefore, the first preset window length is 100 ms. When the electronic device travels at the slowest speed of 0.1 m / s, the delayed displacement caused by detecting a collision is 1 cm.

[0132] In S423, determining the collision evaluation information corresponding to the first preset window length according to the plurality of data points of the collision sensor includes: calculating the collision evaluation information of the plurality of data points within the first preset window length according to a preset evaluation formula.

[0133] As mentioned above, the collision evaluation information may include signal energy, and the signal energy is the total energy transmitted by the target signal within a time equal to the first preset window length.

[0134] When the collision evaluation information includes signal energy, determining the collision evaluation information corresponding to the first preset window length according to the plurality of data points of the collision sensor includes the following steps: determining the data points that fall within the first preset window length as target data points according to the plurality of data points of the collision sensor, and performing energy integration on the plurality of target data points according to the first preset window length to obtain the signal energy.

[0135] For example, in this embodiment, the signal energy is calculated according to a preset energy formula, and the preset energy formula is as follows: where E x is the signal energy, |x(t)| is the amplitude of the target data point, and t2 - t1 is the first preset window length.

[0136] In some embodiments, the collision evaluation information includes signal peak value, and the signal peak value is the maximum amplitude of the target signal within the first preset window length.

[0137] When the collision evaluation information includes a signal peak value, determining the collision evaluation information corresponding to the first preset window length includes the following steps: Based on multiple data points of the collision sensor, determine the data points falling within the first preset window length as target data points, and screen out the maximum amplitude among the multiple target data points as the signal peak value.

[0138] In some embodiments, the electronic device includes a collision part, and at least two collision sensors are installed at intervals along the circumferential direction of the collision part. Each collision sensor is configured with a detection area on the collision part.

[0139] Please refer to Figure 5b , the collision part 11 is provided with 4 detection areas. As Figure 6 shown, expand the collision part 11 to obtain the expanded collision part. The 4 detection areas are respectively the first detection area 51, the second detection area 52, the third detection area 53 and the fourth detection area 54. One or more than two collision sensors can be set in each detection area. The second detection area 52 and the third detection area 53 are distributed on both sides of the forward direction of the electronic device for detecting collisions occurring in front of the electronic device. Among them, the collision sensor pkt2 in the second detection area 52 or the collision sensor pkt3 in the third detection area 53 is centered. The first detection area 51 and the fourth detection area 54 are distributed on both sides of the side edge of the electronic device for detecting collisions occurring on the side of the electronic device. Among them, the collision sensor pkt1 is set at a position in the first detection area 51 close to the second detection area 52, and the collision sensor pkt4 is set at a position in the fourth detection area 54 close to the third detection area 53.

[0140] It can be understood that collisions are likely to occur in the second detection area 52 and the third detection area 53, and are not likely to occur in the first detection area 51 and the fourth detection area 54.

[0141] It can be understood that the area of each detection area can be the same or different. As Figure 6 shown, the second detection area 52 and the third detection area 53 are the same, and the first detection area 51 and the fourth detection area 54 are the same. In order to realize the 360-degree collision perception of the electronic device as much as possible, the third detection area 53 and the fourth detection area 54 will extend backward to near the back of the electronic device, and the area of the third detection area 53 is larger than the area of the second detection area 52.

[0142] It can also be understood that those skilled in the art can customize the detection area according to the structure of the collision part, and this is not limited to the way provided in this embodiment.

[0143] It can also be understood that the number of detection areas can be customized by the designer according to the structure of the collision part and service requirements, and this is not limited here.

[0144] Determining a target collision area based on the collision evaluation information of each collision sensor includes: according to the collision evaluation information of each collision sensor, determining at least one detection partition that satisfies the first maximum likelihood condition in multiple detection partitions as the target detection partition, where the target detection partition is the target collision area. In this embodiment, by configuring a corresponding detection partition for each collision sensor and dividing the collision part into multiple detection partitions, all areas of the collision part can be covered in a planned manner, which is beneficial to detecting the collision positions occurring on the collision part without omission and in all directions.

[0145] The collision evaluation information includes a first type of collision evaluation value, such as the first type of collision evaluation value being signal energy or signal peak value or zero-crossing rate. Determining at least one detection partition that satisfies the first maximum likelihood condition in multiple detection partitions according to the collision evaluation information of each collision sensor includes:

[0146] S431: Select two first type of collision evaluation values with the largest numerical values from among the first type of collision evaluation values to obtain a first evaluation value and a second evaluation value, where the first evaluation value is greater than the second evaluation value.

[0147] S432: Determine whether the detection partition corresponding to the first evaluation value is adjacent to the detection partition corresponding to the second evaluation value.

[0148] S433: If they are adjacent, determine that both the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value satisfy the first maximum likelihood condition and are both target detection partitions.

[0149] In S431, selecting two first type of collision evaluation values with the largest numerical values from among the first type of collision evaluation values to obtain a first evaluation value and a second evaluation value includes the following steps: sorting the first type of collision evaluation values according to a first preset sorting order to obtain two first type of collision evaluation values with the largest numerical values, and the two first type of collision evaluation values with the largest numerical values are respectively the first evaluation value and the second evaluation value. Among them, the first preset sorting order is the order from large to small or the order from small to large.

[0150] In S432, determining whether the detection partition corresponding to the first evaluation value is adjacent to the detection partition corresponding to the second evaluation value includes: according to the first preset sorting order, configuring a sorting index for the detection partition corresponding to each first type of collision evaluation value, determining the sorting index corresponding to the first evaluation value as the first sorting index, determining the sorting index corresponding to the second evaluation value as the second sorting index, calculating the absolute value of the difference between the first sorting index and the second sorting index, and determining whether the absolute value of the difference is the natural number 1.

[0151] In S433, determining that both the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are target detection partitions includes: if the absolute value of the difference is a natural number, the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are adjacent, the detection partition corresponding to the first evaluation value is set as the first reference partition, and the detection partition corresponding to the second evaluation value is set as the second reference partition; if the absolute value of the difference is not a natural number, the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are not adjacent.

[0152] For example, the first type of collision evaluation value is signal energy. In this embodiment, the signal energies of each collision sensor are obtained, where the signal energy is Ej, k is the number of detection partitions, and j ∈ (1, 2,..., k). In this embodiment, the signal energies are sorted in descending order, and a sorting index is configured for the detection partition corresponding to each signal energy, which are m1, m2, m3,..., mk respectively. Among them, the first sorting index corresponding to the first evaluation value is m1, the second sorting index corresponding to the second evaluation value is m2, and the absolute value of the difference between the first sorting index and the second sorting index is |m1 - m2|. If |m1 - m2| = 1, it means that the first sorting index and the second sorting index are consecutive integers, that is, the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are adjacent, the detection partition corresponding to the first evaluation value is set as the first reference partition r1, and the detection partition corresponding to the second evaluation value is set as the second reference partition r2. If |m1 - m2| is not equal to 1, it means that the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are not adjacent.

[0153] In some embodiments, the collision evaluation information further includes a second type of collision evaluation value. For example, when the first type of collision evaluation value is signal energy, the second type of collision evaluation value is signal peak value, or when the first type of collision evaluation value is signal peak value, the second type of collision evaluation value is signal energy.

[0154] If the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value are not adjacent, before determining the collision position according to the target detection partition, the method further includes the following steps:

[0155] S46: Obtain the second type of collision evaluation values of each collision sensor.

[0156] S47: Select the two second type of collision evaluation values with the largest numerical values from the second type of collision evaluation values of each collision sensor to obtain a third evaluation value and a fourth evaluation value, and the third evaluation value is greater than the fourth evaluation value.

[0157] S48: Determine whether the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value are adjacent.

[0158] S49: If they are adjacent, the detection partitions corresponding to the third evaluation value and the fourth evaluation value both meet the preset constraint conditions and are both target detection partitions.

[0159] In S46, in this embodiment, the target signals of each collision sensor are traversed to determine the signal peak values of each collision sensor.

[0160] In S47, screening out the two second - type collision evaluation values with the largest numerical values from the second - type collision evaluation values to obtain the third evaluation value and the fourth evaluation value includes: sorting the second - type collision evaluation values according to the first preset sorting order to obtain the two second - type collision evaluation values with the largest numerical values, and the two second - type collision evaluation values with the largest numerical values are respectively the third evaluation value and the fourth evaluation value.

[0161] In S48, determining whether the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value are adjacent includes: according to the first preset sorting order, configuring a sorting index for the detection partition corresponding to each second - type collision evaluation value, determining the sorting index corresponding to the third evaluation value as the first sorting index, determining the sorting index corresponding to the fourth evaluation value as the second sorting index, calculating the absolute value of the difference between the first sorting index and the second sorting index, and determining whether the absolute value of the difference is the natural number 1.

[0162] In S49, if the absolute value of the difference is the natural number 1, the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value are adjacent, setting the detection partition corresponding to the third evaluation value as the first reference partition, and setting the detection partition corresponding to the fourth evaluation value as the second reference partition. If the absolute value of the difference is not the natural number 1, the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value are not adjacent.

[0163] For example, the second type of collision evaluation value is the signal peak value. In this embodiment, the signal peak values of each collision sensor are obtained, where the signal peak value is Vj, and j ∈ (1, 2,..., k). In this embodiment, the signal peak values are sorted in descending order, and sorting indexes are configured for the detection partitions corresponding to each signal energy, which are n1, n2, n3,..., nk respectively. Among them, the first sorting index corresponding to the third evaluation value is n1, the second sorting index corresponding to the fourth evaluation value is n2, and the absolute value of the difference between the first sorting index and the second sorting index is |n1 - n2|. If |n1 - n2| = 1, it means that the first sorting index and the second sorting index are consecutive integers, that is, the detection partition corresponding to the third evaluation value is adjacent to the detection partition corresponding to the fourth evaluation value. The detection partition corresponding to the third evaluation value is set as the first reference partition r1, and the detection partition corresponding to the second evaluation value is set as the second reference partition r2. If |n1 - n2| is not equal to 1, it means that the detection partition corresponding to the third evaluation value is not adjacent to the detection partition corresponding to the fourth evaluation value.

[0164] When there is an error in the signal energy, this situation is likely to result in the inability to find two adjacent detection partitions. However, in this embodiment, in the above situation, the signal peak value is combined again, and the signal peak value is used to perform a search operation again in order to find two adjacent detection partitions, which is beneficial to accurately and reliably detecting the collision position.

[0165] In some embodiments, if the detection partition corresponding to the third evaluation value is not adjacent to the detection partition corresponding to the fourth evaluation value, the detection partition corresponding to the first evaluation value is determined as the target detection partition. As described above, if |n1 - n2| is not equal to 1, it means that the detection partition corresponding to the third evaluation value is not adjacent to the detection partition corresponding to the fourth evaluation value, indicating that the collision is mainly concentrated in the detection partition corresponding to the first evaluation value. Therefore, in this embodiment, the detection partition corresponding to the first evaluation value is set as the target detection partition.

[0166] In some embodiments, the detection partition includes a first detection sub-region and at least one second detection sub-region. The second detection sub-region is adjacent to the first detection sub-region, and the collision sensor is disposed on the first detection sub-region.

[0167] Please refer to Figure 7 , the first detection partition 51 is divided into a first detection sub-region a11 and a second detection sub-region a12. Among them, the first detection sub-region a11 covers the position of the collision sensor pkt1, that is, the collision sensor pkt1 is disposed on the first detection sub-region a11. There is a large partition boundary line 101 between the first detection partition 51 and the second detection partition 52, and both the first detection partition 51 and the second detection partition 52 are provided with small partition boundary lines. Among them, the small partition boundary line 102 of the second detection sub-region a12 coincides with the large partition boundary line 101.

[0168] The area of the first detection sub-region a11 far from the second detection sub-region a12 is the rear-right area of the electronic device far from the due front. Collisions are not likely to occur in the rear-right area. Collisions of the electronic device are likely to occur in the due front area of the electronic device and the left and right areas deviating from the due front area. In this embodiment, the collision sensors are not arranged in the rear-right area in the manner of equal-area division, and the change in the collision signal at any position in the rear-right area is not obvious. Therefore, in this embodiment, the area of the first detection sub-region a11 can be set to be larger than the area of the second detection sub-region a12, so as to adapt to the collision part structure of the electronic device and reduce the layout of the collision sensors in the rear-right area.

[0169] The second detection partition 52 is evenly divided into three equal detection sub-regions, namely the second detection sub-region a21, the first detection sub-region a22, and the second detection sub-region a23. Among them, the first detection sub-region a22 covers the position of the collision sensor pkt2. The second detection sub-region a21 and the second detection sub-region a23 are located on the left and right sides of the first detection sub-region a22.

[0170] The third detection partition 53 is evenly divided into three equal detection sub-regions, namely the second detection sub-region a32, the first detection sub-region a33, and the second detection sub-region a34. Among them, the first detection sub-region a33 covers the position of the collision sensor pkt3. The second detection sub-region a32 and the second detection sub-region a34 are located on the left and right sides of the first detection sub-region a33.

[0171] The fourth detection partition 54 is divided into the first detection sub-region a44 and the second detection sub-region a43. Among them, the first detection sub-region a44 covers the position of the collision sensor pkt4. Based on the same reasons as the first detection sub-region a11 and the second detection sub-region a12 of the first detection partition 511, in this embodiment, the area of the first detection sub-region a44 can be set to be larger than the area of the second detection sub-region a43.

[0172] In some embodiments, determining the collision position according to the target collision area includes the following steps:

[0173] S461: Determine the detection sub-region that satisfies the second maximum likelihood condition of a collision in at least one target detection partition as the reference detection sub-region.

[0174] S462: Determine the collision position according to the position of the reference detection sub-region in the collision part.

[0175] In S461, the second maximum likelihood condition is used to screen out the detection sub-region corresponding to the maximum probability of a collision in the target detection partition. Those skilled in the art can reasonably set the maximum likelihood condition according to the content disclosed in this embodiment.

[0176] In some embodiments, the second maximum likelihood condition is: the detection sub-region corresponding to the maximum first-type collision evaluation value or the maximum second-type collision evaluation value. The detection sub-region that satisfies the second maximum likelihood condition for a collision is the detection sub-region corresponding to the maximum collision evaluation value.

[0177] In some embodiments, the second maximum likelihood condition is: when the ratio of the first-type collision evaluation value of the first reference partition to the first-type collision evaluation value of the second reference partition is less than a preset ratio threshold, the second detection sub-region adjacent to the second reference partition in the first reference partition belongs to the detection sub-region corresponding to the maximum probability of a collision, and the detection sub-region that satisfies the second maximum likelihood condition for a collision is the second detection sub-region of the first reference partition. Or, when the ratio of the first-type collision evaluation value of the first reference partition to the first-type collision evaluation value of the second reference partition is not less than the preset ratio threshold, the first detection sub-region of the first reference partition belongs to the detection sub-region corresponding to the maximum probability of a collision, and the detection sub-region that satisfies the second maximum likelihood condition for a collision is the first detection sub-region of the first reference partition.

[0178] In some embodiments, the second maximum likelihood condition is: when the ratio of the second-type collision evaluation value of the first reference partition to the second-type collision evaluation value of the second reference partition is less than a preset ratio threshold, the second detection sub-region adjacent to the second reference partition in the first reference partition belongs to the detection sub-region corresponding to the maximum probability of a collision, and the detection sub-region that satisfies the second maximum likelihood condition for a collision is the second detection sub-region of the first reference partition. Or, when the ratio of the second-type collision evaluation value of the first reference partition to the second-type collision evaluation value of the second reference partition is not less than the preset ratio threshold, the first detection sub-region of the first reference partition belongs to the detection sub-region corresponding to the maximum probability of a collision, and the detection sub-region that satisfies the second maximum likelihood condition for a collision is the first detection sub-region of the first reference partition.

[0179] Determining the detection sub-region that satisfies the second maximum likelihood condition for a collision in at least one target detection partition as the reference detection sub-region includes the following steps:

[0180] S4611: If the number of target detection partitions is at least two, calculate the ratio of the first type of collision evaluation value of the first reference partition to the first type of collision evaluation value of the second reference partition. The first reference partition is the target detection partition with the maximum first type of collision evaluation value among at least one target detection partition, and the second reference partition is the target detection partition in at least one target detection partition where the first type of collision evaluation value is smaller than that of the first reference partition but larger than that of the remaining target detection partitions. Or, the first reference partition is the target detection partition with the maximum second type of collision evaluation value among at least one target detection partition, and the second reference partition is the target detection partition in at least one target detection partition where the second type of collision evaluation value is smaller than that of the first reference partition but larger than that of the remaining target detection partitions.

[0181] S4612: If the ratio is less than the preset ratio threshold, determine that the second detection sub-region adjacent to the second reference partition in the first reference partition satisfies the maximum likelihood condition of a collision, and the second detection sub-region is the reference detection sub-region.

[0182] S4613: If the ratio is not less than the preset ratio threshold, determine the maximum likelihood condition of a collision for the first detection sub-region of the first reference partition, and the first detection sub-region is the reference detection sub-region.

[0183] S4614: If the number of target detection partitions is 1, determine the maximum likelihood condition of a collision for the first detection sub-region of the first reference partition, and the first detection sub-region is the reference detection sub-region.

[0184] For example, the number of target detection partitions is 2. If the first detection partition 51 is the first reference partition and the second detection partition 52 is the second reference partition. When the ratio of the first type of collision evaluation value of the first detection partition 51 to the first type of collision evaluation value of the second detection partition 52 is less than the preset ratio threshold, since the second detection sub-region adjacent to the second reference partition in the first reference partition is determined to be the second detection sub-region a12, and the second detection sub-region a12 is the reference detection sub-region. Since the position of the second detection sub-region a12 at the collision part is the front right, therefore, the collision position is the front right information.

[0185] When the ratio of the first type of collision evaluation value of the first detection partition 51 to the first type of collision evaluation value of the second detection partition 52 is greater than the preset ratio threshold, the first detection sub-region a11 of the first detection partition 51 is the reference detection sub-region. Since the position of the first detection sub-region a11 at the collision part is the rear right, therefore, the collision position is the rear right information.

[0186] For another example, the number of target detection partitions is 2. If the second detection partition 52 is the first reference partition and the third detection partition 53 is the second reference partition. When the ratio of the first type of collision evaluation value of the second detection partition 52 to the first type of collision evaluation value of the third detection partition 53 is less than the preset ratio threshold, since the second detection sub-region adjacent to the second reference partition is determined as the second detection sub-region a23 in the first reference partition, and the second detection sub-region a23 is the reference detection sub-region. Since the position of the second detection sub-region a23 at the collision part is rightward and forward, the collision position is the rightward and forward information.

[0187] When the ratio of the first type of collision evaluation value of the second detection partition 52 to the first type of collision evaluation value of the third detection partition 53 is greater than the preset ratio threshold, the first detection sub-region a22 of the second detection partition 52 is the reference detection sub-region. Since the position of the first detection sub-region a22 at the collision part is middle and forward, the collision position is the middle and forward information.

[0188] For another example, the number of target detection partitions is 1, where the fourth detection partition 54 is the target detection partition and the first detection sub-region a44 is the reference detection sub-region. Since the position of the first detection sub-region a44 at the collision part is left rear, the collision position is the left rear information.

[0189] This embodiment can not only generate the collision position according to the target detection partition with a larger area, but also further determine the detection sub-region where the collision position occurs on the basis of the target detection partition with a larger area, so that the collision position and collision direction can be determined more granularly, which is beneficial for the electronic device to reliably and accurately perform the obstacle avoidance operation.

[0190] In some embodiments, before determining the target detection partition, the collision evaluation information includes the first type of collision evaluation value and the second type of collision evaluation value, and the method for determining the collision position further includes the following steps:

[0191] S410: Calculate the sum of the first type of collision evaluation values of each collision sensor to obtain the total collision evaluation value.

[0192] S411: Screen out the maximum second type of collision evaluation value among the second type of collision evaluation values of each collision sensor.

[0193] S412: Generate collision effect information according to the maximum second type of collision evaluation value and the total collision evaluation value.

[0194] S413: Control whether the electronic device continues to perform the collision detection operation according to the collision effect information.

[0195] In S410, this embodiment calculates the total collision evaluation value according to the following formula: the total collision evaluation value i is the serial number of the collision sensor, Ei is the first - type collision evaluation value of the i - th collision sensor, k is the total number of collision sensors, and Eall is the total collision evaluation value.

[0196] In S411, in this embodiment, the maximum second - type collision evaluation value can be screened out according to the method of the above - mentioned embodiment, which will not be elaborated here.

[0197] In S412, the collision effect information includes collision valid information and collision invalid information. The collision valid information is used to indicate that the collision that occurs to the electronic device is valid, and the collision invalid information is used to indicate that the collision that occurs to the electronic device is invalid. Generating the collision effect information according to the maximum second - type collision evaluation value and the total collision evaluation value includes: determining whether the total collision evaluation value is greater than the first preset evaluation threshold and whether the maximum second - type collision evaluation value is greater than the second preset evaluation threshold. If both are greater, then determine that the collision effect information is collision valid information; if not, then determine that the collision effect information is collision invalid information.

[0198] In S413, controlling whether the electronic device continues to perform the collision detection operation according to the collision effect information includes: if the collision effect information is collision valid information, then enter the step of determining the target detection area; if the collision effect information is collision invalid information, then stop the collision detection operation.

[0199] For example, the total collision evaluation value of each collision sensor In this embodiment, the second - type collision evaluation values of each collision sensor are sorted in descending order to obtain the maximum second - type collision evaluation value VPmax. This embodiment determines whether the total collision evaluation value E all is greater than the first preset evaluation threshold Eth and whether the maximum second - type collision evaluation value VPmax is greater than the second preset evaluation threshold VPth. If both are greater, then determine that the electronic device has had a collision; if not, then determine that the electronic device has not had a collision. At this time, the collision signal may be an extremely light rub or an interference signal of instantaneous noise. By judging the validity of the collision before performing the collision detection operation in this embodiment, misjudgment can be avoided and the collision detection efficiency can also be improved.

[0200] To elaborate in detail the process of determining the collision position in this embodiment, this embodiment will be elaborated in detail in combination with Figure 8 wherein, in this embodiment, the signal energy is used as the first - type collision evaluation value, and the signal peak value is used as the second - type collision evaluation value, specifically as follows:

[0201] S111: Calculate the sum of the signal energies of each collision sensor to obtain the total energy Eall.

[0202] S112: Screen out the maximum signal peak value VPmax from the signal peak values of each collision sensor.

[0203] S113: Determine whether Eall is greater than Eth and whether VPmax is greater than VPth. If both are greater, execute S114; if not, execute S125.

[0204] S114: If Eall is greater than Eth and VPmax is greater than VPth, determine that the collision effectiveness information is collision valid information and enter the execution step S115.

[0205] S115: Configure indexes n1 to nk for the signal peaks of each collision sensor in descending order and enter the execution step S116.

[0206] S116: Configure indexes m1 to mk for the signal energies of each collision sensor in descending order and enter the execution step S117.

[0207] S117: Determine whether │m1 - m2│ == 1? If so, execute step S118; if not, execute step S122.

[0208] S118: Determine that the detection partition corresponding to index m1 is the target detection partition r1, and the detection partition corresponding to index m2 is the target detection partition r2, and enter the execution step S119.

[0209] S119: Determine whether gEr1 / gEr2 < Rth. If so, execute step S120; if not, execute step S121.

[0210] S120: Determine that the position of the second detection sub - area of the target detection partition r1 in the collision part is the collision position.

[0211] S121: Determine that the position of the first detection sub - area of the target detection partition r1 in the collision part is the collision position.

[0212] S122: Determine whether │n1 - n2│ == 1? If so, execute step S123; if not, execute step S124.

[0213] S123: Determine that the detection partition corresponding to index n1 is the target detection partition r1, and the detection partition corresponding to index n2 is the target detection partition r2, and enter the execution step S119.

[0214] S124: Determine that the position of the first detection sub - area of the target detection partition r1 in the collision part is the collision position.

[0215] S125: End.

[0216] Generally speaking, this embodiment can not only eliminate the influence of signal crosstalk by virtue of the characteristics of the target signal, ensuring reliable detection of the collision position. In addition, this embodiment can divide the collision part into large-area detection partitions, and further divide them into small-area detection sub-areas on the basis of the detection partitions, so as to detect the collision position more granularly and accurately, which is conducive to the electronic device to reliably and accurately take appropriate obstacle avoidance measures.

[0217] In some embodiments, before determining the collision evaluation information of the collision sensor, the method further includes the following steps:

[0218] S51: Generate collision detection information according to the target signal, where the collision detection information includes collision presence information or collision absence information.

[0219] S52: If the collision detection information is collision presence information, enter the step of determining the collision evaluation information of the collision sensor.

[0220] S53: If the collision detection information is collision absence information, end the operation of determining the collision position.

[0221] In S51, the collision detection information includes collision presence information and collision absence information. The collision presence information is used to indicate that the electronic device has a collision, and the collision absence information is used to indicate that the electronic device does not have a collision. In some embodiments, this embodiment can process the target signal using a preset signal energy algorithm to obtain the collision detection information. In some embodiments, this embodiment can process the target signal using a preset signal peak algorithm to obtain the collision detection information. In some embodiments, this embodiment can process the target signal using a preset signal zero-crossing rate algorithm to obtain the collision detection information. In some embodiments, this embodiment can process the target signal using the collision detection algorithm provided by this embodiment to obtain the collision detection information.

[0222] In S52, if the collision detection information is collision presence information, it indicates that the electronic device has a collision. Therefore, the electronic device can further determine the collision position, and then generate the collision evaluation information of the collision sensor.

[0223] In S53, if the collision detection information is collision absence information, it indicates that the electronic device does not have a collision. Therefore, this embodiment ends the operation of determining the collision position. This embodiment can determine whether to perform the operation of determining the collision position according to the detection result by detecting whether the electronic device has a collision, without frequently detecting the collision position, which is conducive to improving the efficiency of determining the collision position.

[0224] In some embodiments, generating collision detection information according to the target signal includes:

[0225] S511: Collect multiple data points on the target signal according to a first preset window length.

[0226] S512: Perform a trigger edge detection operation based on the multiple data points to obtain trigger edge detection information.

[0227] S513: Generate collision detection information based on the trigger edge detection information.

[0228] In S511, the first preset window length is less than the collision duration of the target signal, and the collision duration is the difference between the start time and the end time of the target signal.

[0229] Please refer to Figure 9 , the initial amplitude of the collision sensor when no collision is detected is 34200. At time point t1 = 14 ms, the electronic device collides with an obstacle, and the collision sensor is squeezed by the obstacle to generate a collision signal, which can be processed to obtain the target signal. Among them, the target signal starts to rise rapidly from 14 ms. This process is from the moment of contacting the obstacle to the moment of deep contact with the obstacle. Since the thrust generated during this collision process is unstable, the collision sensor is squeezed to slowly accumulate positive charges, and the voltage of the target signal gradually increases. When the thrust is stable, the charge amount of the collision sensor reaches the maximum, and the voltage of the target signal reaches the peak at 25 ms. Therefore, the voltage of the target signal reaches the peak.

[0230] After 25 ms, the collision sensor starts to discharge, and the voltage of the target signal gradually decreases. At 34 ms, the voltage of the target signal drops to the preset bias voltage. Due to the device reason of the collision sensor, the voltage of the target signal will continue to decrease. When the voltage of the target signal drops to 38 ms, the voltage of the target signal starts to rise. At time point t2 = 42 ms, the target signal reaches the preset bias voltage again. After 42 ms, the voltage of the target signal oscillates slightly and finally converges to the preset bias voltage. However, since the amplitude of the signal after 42 ms is relatively small, the signal after 42 ms is usually not used as the target signal.

[0231] As Figure 9 shown, the start time of the target signal is time point t1, the end time is time point t2, and the collision duration is the difference between time point t2 and time point t1.

[0232] Generally, the collision duration of the target signal is usually greater than 200 ms. When performing collision detection in related technologies, methods such as the VPP threshold judgment method or the energy judgment method are usually adopted. Such methods usually require a target signal with a complete collision duration to perform collision detection. As mentioned above, the time required to perform collision detection based on a complete collision signal is usually greater than 400 ms. In this embodiment, multiple data points are collected on the target signal by using a first preset window length for collision detection. Since the first preset window length is less than the collision duration of the target signal, the time for performing collision detection in this embodiment is relatively short.

[0233] In S512, the trigger edges include the rising edge and the falling edge. Usually, the collision signal will include the rising edge and the falling edge, and the target signal will also include the rising edge and the falling edge. When the electronic device collides, the voltage of the collision signal collected by the collision sensor will increase or decrease sharply based on the preset bias voltage, thus resulting in a rising edge or a falling edge. This embodiment can give a collision detection result by detecting the rising edge or the falling edge. Among them, when the analog signal is transferred to the digital signal, the voltage of the collision signal is likely to appear negative, which is not conducive to subsequent calculations. Therefore, in this embodiment, the preset bias voltage is shifted from 0 to the corresponding positive number. When the electronic device does not collide, the collision signal collected by the collision sensor is relatively stable, and there is usually no signal in the target frequency band. The collision signal will not increase or decrease steeply, that is, there will be no rising edge or falling edge.

[0234] The trigger edge detection operation includes the rising edge detection operation and the falling edge detection operation. The trigger edge detection information includes the trigger edge prompt information and the non-trigger edge information. The trigger edge prompt information is used to indicate that multiple data points meet the morphological requirements of the trigger edge, and the non-trigger edge information is used to indicate that multiple data points do not meet the change trend of the trigger edge. The trigger edge prompt information includes the rising edge existence result and the falling edge existence result. The rising edge existence result is used to indicate that multiple data points meet the change trend of the rising edge, and the falling edge existence result is used to indicate that multiple data points meet the change trend of the falling edge.

[0235] It can be understood that this embodiment can perform the rising edge detection operation to detect collisions, or perform the falling edge detection operation to detect collisions, or perform the above two trigger edge detection operations.

[0236] It can also be understood that since this embodiment only needs to perform the rising edge detection operation or the falling edge detection operation, this embodiment does not require a target signal with a complete collision duration and can also achieve the purpose of collision detection.

[0237] In S513, generating collision detection information based on trigger edge detection information includes: if the trigger edge detection information is trigger edge prompt information, generating collision existence information; if the trigger edge detection information is non-trigger edge information, generating collision non-existence information.

[0238] In this embodiment, it is not necessary to take the target signal of the complete collision duration for collision detection. Instead, a short window method is used to collect multiple data points in the target signal for collision detection. Since the first preset window length of the short window method is less than the collision duration of the target signal, this embodiment can perform collision detection faster, shorten the collision response time, and has a relatively high detection efficiency. In addition, this embodiment uses the trigger edge detection method for collision detection. Compared with the related technology that needs to calculate the sum of the energies of the entire window, the calculation efficiency of this embodiment is high, which is beneficial to improving the speed and efficiency of collision detection.

[0239] In some embodiments, multiple data points can be sequentially divided into multiple data groups, and each data group includes at least one data point. In this embodiment, multiple data points can be sequentially divided into multiple data groups according to a preset quantity. Among them, each data group includes the same number of data points as the preset quantity.

[0240] In some embodiments, the preset quantity is 1. That is, each data group contains 1 data point. For example, the data point set Q of multiple data points = {q1, q2,..., q i ,..., q m}, where q i is the i-th data point, and the data point set Q is divided into m data groups, and each data group contains one data point.

[0241] In some embodiments, the preset quantity is a, and a is an integer greater than 1. The data point set Q is divided into b data groups, where b = m / a, and the result of m divided by a is determined according to the ceiling method. Each data group contains b data points, such as b being 2 or 3 or 4, etc.

[0242] Performing a trigger edge detection operation based on multiple data points to obtain trigger edge detection information includes the following steps:

[0243] S5121: Calculating a differential signal according to the amplitudes of two adjacent data groups.

[0244] S5122: Generating trigger edge detection information according to multiple differential signals.

[0245] In S5121, in some embodiments, when the data group includes one data point, the amplitude of the data group is the amplitude of the data point. In some embodiments, when the data group includes at least two data points, the amplitude of the data group is the average amplitude of all the data points in the data group. Since this embodiment uses the average amplitude of all the data points in the data group as the amplitude of the data group, it can reduce the influence caused by the deviation of the amplitude of individual data points. That is to say, this method can smooth the amplitude of the data group, making the amplitude of the data group more reliable.

[0246] In some embodiments, calculating the differential signal according to the amplitudes of two adjacent data groups includes: calculating the difference between the amplitudes of two adjacent data groups to obtain the differential signal. For example, the differential signal ΔG j-1 = G j - G j-1 , where G j is the amplitude of the jth data group. According to the amplitudes of multiple data groups in this embodiment, the following differential signals can be obtained: ΔG1, ΔG2, ΔG3.... Then, trigger edge detection information is generated according to multiple differential signals.

[0247] In some embodiments, calculating the differential signal according to the amplitudes of two adjacent data groups includes: respectively processing the amplitudes of two adjacent data groups according to a preset weighting algorithm to obtain a first weighted result and a second weighted result, and subtracting the first weighted result from the second weighted result to obtain the differential signal.

[0248] Since effective trigger edges are all continuously increasing significantly or continuously decreasing significantly, this embodiment uses the differential signal to characterize the situation of "continuously increasing significantly" or "continuously decreasing significantly", and combines the short window detection method, without obtaining the target signal of the complete collision duration, so as to be able to quickly and effectively perform collision detection.

[0249] In S5122, generating trigger edge detection information according to multiple differential signals includes the following steps:

[0250] S51221: Determine whether the absolute values of M differential signals are greater than a preset differential threshold, where the number of all differential signals is N, both M and N are positive integers, and M ≤ N.

[0251] S51222: If it is greater, determine whether the M differential signals meet the preset monotonic change condition. If they meet, generate trigger edge prompt information.

[0252] S51223: If they do not meet or if it is less, generate non-trigger edge information.

[0253] In S51221, M can be less than N, or M can be equal to N. For example, if the number of all differential signals is 4, M can be 2 or 3, or M can be 4. The preset differential threshold is customized by the designer according to engineering experience.

[0254] In some embodiments, M = N. Determining whether the absolute values of M differential signals are greater than a preset differential threshold includes the following steps: determining whether the absolute values of all differential signals are greater than the preset differential threshold.

[0255] In some embodiments, M is less than N. Determining whether the absolute values of M differential signals are greater than a preset differential threshold includes the following steps: selecting M differential signals with continuous time among all differential signals, and determining whether the absolute values of the M differential signals are greater than the preset differential threshold.

[0256] The change trend of the rising edge or the falling edge is reflected in a large amplitude change between two adjacent data points, that is, the absolute value of the differential signal needs to be greater than the preset differential threshold. If the absolute values of multiple differential signals are all greater than the preset differential threshold, it means that the amplitude of the data points corresponding to the M differential signals fluctuates greatly, which meets a necessary condition for the trigger edge. In this embodiment, it is further determined whether a trigger edge appears for the M differential signals. If the absolute values of multiple differential signals are not all greater than the preset differential threshold, it means that the amplitude of the data points corresponding to the M differential signals fluctuates little, which does not meet a necessary condition for the trigger edge. In this embodiment, it is not necessary to further determine whether a trigger edge appears for the M differential signals, thus saving computing power, and in this way, the trigger edge detection efficiency can be improved.

[0257] In S51222 and S51223, the preset monotonic change condition includes a preset monotonic increasing condition or a preset monotonic decreasing condition. In some embodiments, determining whether M differential signals meet the preset monotonic change condition includes: determining whether the M differential signals meet the preset monotonic increasing condition, and / or determining whether the M differential signals meet the preset monotonic decreasing condition.

[0258] In some embodiments, M = N. Determining whether M differential signals meet the preset monotonic change condition includes: determining whether all differential signals meet the preset monotonic increasing condition, and / or determining whether all differential signals meet the preset monotonic decreasing condition.

[0259] In some embodiments, determining whether M differential signals satisfy a preset monotonically increasing condition includes: determining whether all M differential signals are greater than a preset positive threshold or determining whether all differential signals are greater than the preset positive threshold. If all are greater, it is determined that the M differential signals satisfy the preset monotonically increasing condition or it is determined that all differential signals satisfy the preset monotonically increasing condition. If not all are greater, it is determined that the M differential signals do not satisfy the preset monotonically increasing condition or it is determined that all differential signals do not satisfy the preset monotonically increasing condition. If the M differential signals satisfy the preset monotonically increasing condition or all differential signals satisfy the preset monotonically increasing condition, it indicates that the change trend of the M data points or all data points is an upward edge change trend.

[0260] In some embodiments, determining whether M differential signals satisfy a preset monotonically decreasing condition includes: determining whether all M differential signals are less than a preset negative threshold or determining whether all differential signals are greater than the preset negative threshold. If all are less, it is determined that the M differential signals satisfy the preset monotonically decreasing condition or it is determined that all differential signals satisfy the preset monotonically decreasing condition. If not all are less, it is determined that the M differential signals do not satisfy the preset monotonically decreasing condition. If multiple differential signals satisfy the preset monotonically decreasing condition or all differential signals satisfy the preset monotonically decreasing condition, it indicates that the change trend of the M data points or all data points is a downward edge change trend.

[0261] It can be understood that regardless of whether the change trend of multiple data points is an upward edge change trend or a downward edge change trend, in this embodiment, it is determined that the trigger edge detection information is the trigger edge prompt information, and the collision existence information is generated accordingly. This embodiment adopts a short window method and combines differential signals to detect the upward edge or the downward edge, so as to be able to detect collisions reliably and quickly and reduce the delay of collision detection.

[0262] In some embodiments, before extracting multiple data points, the method for determining the collision position further includes the following steps:

[0263] S54: Obtain the moving speed of the electronic device.

[0264] S55: Determine whether the moving speed is greater than a preset speed threshold.

[0265] S56: If it is greater, extract multiple data points from the target signal according to a first preset window length.

[0266] S57: If it is less, perform a collection operation on the target signal according to a second preset window length to obtain a collection signal, and process the collection signal according to a preset collision detection algorithm to obtain collision detection information, where the second preset window length is greater than the collision duration of the target signal.

[0267] In S54, the moving speed is the speed at which the electronic device travels. The electronic device is provided with a speed sensor, which is used to detect the moving speed of the electronic device.

[0268] In S55, the preset speed threshold is customized by the designer according to engineering experience. For example, the preset speed threshold is 0.1 m / s.

[0269] In S56, when the moving speed is greater than the preset speed threshold, it indicates that the moving speed of the electronic device is relatively fast. In this embodiment, the short-window differential detection method provided by the above various embodiments is used for collision detection, so that the delay distance generated by collision detection can be reduced.

[0270] In S57, when the moving speed is less than the preset speed threshold, it indicates that the moving speed of the electronic device is relatively slow. In this embodiment, the second preset window length can be used to obtain the acquisition signal, and then the preset collision detection algorithm is combined to process the acquisition signal for collision detection. Among them, the preset collision detection algorithm includes the VPP threshold judgment algorithm or the energy judgment algorithm, etc., and the acquisition signal is the target signal of the complete collision duration.

[0271] This embodiment flexibly selects the collision detection method by combining the moving speed and the preset speed threshold. The short-window differential method is selected for collision detection in the case of a large moving speed, which is beneficial to improving the collision detection efficiency. The large-window method is selected for collision detection in the case of a small moving speed, which is beneficial to improving the reliability of collision detection.

[0272] In some embodiments, the second preset window length is greater than or equal to 2 times the collision duration of the target signal.

[0273] Assume that the collision duration of the target signal is L. Please refer to Figure 10, if the second preset window length is set to L, since the target signal appears randomly, there is only one chance for the target signal to completely fall within the second preset window length. In other cases, the second preset window length cannot collect the complete target signal. Therefore, the probability that the target signal completely falls within the second preset window length is 1 / L. Since L is generally very large, this probability is very low. For example, the target signal consists of data points p0, p1, p2, p3, and p4, and the second preset window length L is the length of 5 data points. The target signal {p0, p1, p2, p3, p4} has only one chance to completely fall within the second preset window length. The situations where the target signal does not completely fall within the second preset window length are as follows: only data point p0 falls within the second preset window length, only data points p0 and p1 fall within the second preset window length, only data points p0, p1, and p2 fall within the second preset window length, and only data points p0, p1, p2, and p3 fall within the second preset window length. Then, the probability that the target signal {p0, p1, p2, p3, p4} can completely fall within the second preset window length is 1 / (4 + 1) = 0.2.

[0274] Please refer to Figure 11 , if the second preset window length is set to 2L, then the chance for the target signal to completely fall within the second preset window length is L + 1. Therefore, the probability that the target signal completely falls within the second preset window length is (L + 1) / 2L ≈ 0.5. Therefore, the longer the second preset window length, the better. However, to reduce latency, in this embodiment, it is only necessary to select the second preset window length to be greater than or equal to 2L. For example, the target signal consists of data points p0, p1, p2, p3, and p4, and the second preset window length 2L is the length of 10 data points. The situations where the target signal {p0, p1, p2, p3, p4} has only one chance to completely fall within the second preset window length are as follows: {*, *, *, *, *, p4, p3, p2, p1, p0}, {*, *, *, *, p4, p3, p2, p1, p0, *}, {*, *, *, p4, p3, p2, p1, p0, *, *}, {*, *, p4, p3, p2, p1, p0, *, *, *}, {*, p4, p3, p2, p1, p0, *, *, *, *}, {p4, p3, p2, p1, p0, *, *, *, *, *}. As can be seen from the above, there are 6 chances for the target signal {p0, p1, p2, p3, p4} to have only one chance to completely fall within the second preset window length. Among them, "*" represents empty data, that is, no data can be collected.

[0275] The situations where the target signal {p0, p1, p2, p3, p4} cannot completely fall within the second preset window length are as follows: {p3, p2, p1, p0, *, *, *, *, *, *,}, {p2, p1, p0, *, *, *, *, *, *, *}, {p1, p0, *, *, *, *, *, *, *, *}, {p0, *, *, *, *, *, *, *, *, *}. As can be seen from the above, there are 4 opportunities for the target signal {p0, p1, p2, p3, p4} not to completely fall within the second preset window length.

[0276] The probability that the target signal {p0, p1, p2, p3, p4} can completely fall within the second preset window length is 6 / 10 = 0.6.

[0277] It should be noted that in the above various embodiments, there is not necessarily a certain order between the above steps. Those of ordinary skill in the art can understand from the description of the embodiments of the present application that in different embodiments, the above steps can have different execution orders, that is, they can be executed in parallel, or they can be executed alternately, etc.

[0278] Please refer to Figure 12 , Figure 12 , which is a schematic circuit structure diagram of a controller provided by an embodiment of the present application. As Figure 12 shown, the controller 120 includes one or more processors 121 and a memory 122. Among them, Figure 12 one processor 121 is taken as an example.

[0279] The processor 121 and the memory 122 can be connected through a bus or other means, Figure 12 and the connection through a bus is taken as an example.

[0280] The memory 122, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the method for determining the collision position in the embodiments of the present application. The processor 121 executes various functional applications and data processing of the collision detection device by running the non-volatile software programs, instructions, and modules stored in the memory 122, that is, realizes the functions of each module or unit of the method for determining the collision position provided in the above method embodiments.

[0281] The memory 122 may include high-speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 122 optionally includes a memory remotely located relative to the processor 121, and these remote memories may be connected to the processor 121 through a network. Examples of the above networks include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0282] The program instructions / modules are stored in the memory 122 and, when executed by the one or more processors 121, perform the method for determining the collision position in any of the above method embodiments.

[0283] The embodiment of the present application also provides a non-volatile computer storage medium having computer-executable instructions, and these computer-executable instructions are executed by one or more processors, for example Figure 12 one of the processors 121, enabling the one or more processors to execute the method for determining the collision position in any of the above method embodiments.

[0284] The embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a controller, the controller is enabled to execute the method for determining the collision position in any one of the above.

[0285] The device or equipment embodiments described above are merely illustrative. The unit modules described as separate components may or may not be physically separated, and the components shown as module units may or may not be physical units, that is, they may be located in one place or distributed to multiple network module units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0286] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the related technology can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0287] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; under the idea of the present application, the technical features in the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order, and there are many other changes in different aspects of the present application as described above. For the sake of brevity, they are not provided in detail; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining a collision position, applied to an electronic device, characterized in that The electronic device includes a collision part and at least two of the collision sensors, and the at least two collision sensors are installed at intervals on the collision part. The method includes: Obtaining a collision signal, which is collected by the collision sensor; Determining collision evaluation information of the collision sensor according to the collision signal; Determining a target collision area according to the collision evaluation information of each collision sensor, where the target collision area is the area on the collision part that satisfies the first maximum likelihood condition for a collision to occur; Determining a collision position according to the target collision area.

2. The method according to claim 1, characterized in that The determining the collision evaluation information of the collision sensor according to the collision signal includes: Determining a collision signal with a frequency in a target frequency band as a target signal according to the collision signal; Collecting a plurality of data points on the target signal according to a first preset window length; Determining collision evaluation information corresponding to the first preset window length according to the plurality of data points of the collision sensor.

3. The method according to claim 2, characterized in that, The target signal is a quasi-static force signal or a static force signal, and the target frequency band is [0Hz, 150Hz].

4. The method according to claim 1, wherein Each collision sensor is configured with a detection partition on the collision part. The determining the target collision area according to the collision evaluation information of each collision sensor includes: Determining at least one detection partition that satisfies the first maximum likelihood condition among a plurality of detection partitions as a target detection partition according to the collision evaluation information of each collision sensor, where the target detection partition is the target collision area.

5. The method according to claim 4, characterized in that The collision evaluation information includes a first type of collision evaluation value. The determining at least one detection partition that satisfies the first maximum likelihood condition among a plurality of detection partitions as a target detection partition according to the collision evaluation information of each collision sensor includes: Selecting two first type of collision evaluation values with the largest numerical values from the first type of collision evaluation values to obtain a first evaluation value and a second evaluation value, where the first evaluation value is greater than the second evaluation value; Judging whether the detection partition corresponding to the first evaluation value is adjacent to the detection partition corresponding to the second evaluation value; If they are adjacent, determining that both the detection partition corresponding to the first evaluation value and the detection partition corresponding to the second evaluation value satisfy the first maximum likelihood condition and are both target detection partitions.

6. The method according to claim 5, characterized in that, The collision evaluation information further includes a second type of collision evaluation value. If the detection partition corresponding to the first evaluation value is not adjacent to the detection partition corresponding to the second evaluation value, before determining the collision position according to the target detection partition, the method further includes: Obtaining the second type of collision evaluation values of each collision sensor; Selecting the two largest second type of collision evaluation values from the second type of collision evaluation values to obtain a third evaluation value and a fourth evaluation value, where the third evaluation value is greater than the fourth evaluation value; Judging whether the detection partition corresponding to the third evaluation value is adjacent to the detection partition corresponding to the fourth evaluation value; If they are adjacent, determining that both the detection partition corresponding to the third evaluation value and the detection partition corresponding to the fourth evaluation value satisfy the first maximum likelihood condition and are both target detection partitions.

7. The method according to claim 6, wherein It further includes: If the detection partition corresponding to the third evaluation value is not adjacent to the detection partition corresponding to the fourth evaluation value, determine the detection partition corresponding to the first evaluation value as the target detection partition.

8. The method according to claim 5 or 6, characterized in that, The detection partition includes a first detection sub-region and at least one second detection sub-region, the second detection sub-region is adjacent to the first detection sub-region, the collision sensor is disposed on the first detection sub-region, and determining the collision position according to the target collision region includes: Determine, in at least one of the target detection partitions, a detection sub-region that satisfies the second maximum likelihood condition for a collision to occur as the reference detection sub-region; Determine the collision position according to the position of the reference detection sub-region on the collision part.

9. The method according to claim 8, characterized in that The determining, in at least one of the target detection partitions, a detection sub-region that satisfies the second maximum likelihood condition for a collision to occur as the reference detection sub-region includes: If the number of the target detection partitions is at least two, calculate a ratio of a first type of collision evaluation value of a first reference partition to a first type of collision evaluation value of a second reference partition, the first reference partition being the target detection partition with the maximum first type of collision evaluation value among at least one of the target detection partitions, the second reference partition being, among at least one of the target detection partitions, a target detection partition whose first type of collision evaluation value is smaller than that of the first reference partition but larger than the first type of collision evaluation values of the remaining target detection partitions, or the first reference partition being the target detection partition with the maximum second type of collision evaluation value among at least one of the target detection partitions, the second reference partition being, among at least one of the target detection partitions, a target detection partition whose second type of collision evaluation value is smaller than that of the first reference partition but larger than the second type of collision evaluation values of the remaining target detection partitions; If the ratio is less than a preset ratio threshold, determine that a second detection sub-region adjacent to the second reference partition in the first reference partition satisfies the second maximum likelihood condition for a collision to occur, and the second detection sub-region is the reference detection sub-region; If the ratio is not less than the preset ratio threshold, determine that the first detection sub-region of the first reference partition satisfies the second maximum likelihood condition for a collision to occur, and the first detection sub-region is the reference detection sub-region.

10. The method according to claim 8, wherein The determining, in at least one of the target detection partitions, a detection sub-region that satisfies the second maximum likelihood condition for a collision to occur as the reference detection sub-region includes: If the number of the target detection partitions is 1, determine that the first detection sub-region of the target detection partition satisfies the second maximum likelihood condition for a collision to occur, and the first detection sub-region of the target detection partition is the reference detection sub-region.

11. The method according to claim 6, characterized in that The first type of collision evaluation value is a signal energy value, and the second type of collision evaluation value is a signal peak value.

12. The method according to claim 4, wherein The collision evaluation information includes a first type of collision evaluation value and a second type of collision evaluation value. Before determining the target collision region, the method further includes: Calculate the sum of the first type of collision evaluation values of each of the collision sensors to obtain a total collision evaluation value; Select the maximum second type of collision evaluation value from among the second type of collision evaluation values of each of the collision sensors; Generate collision effectiveness information according to the maximum second type of collision evaluation value and the total collision evaluation value; Control whether the electronic device continues to perform the collision detection operation according to the collision effect information.

13. The method according to claim 12, wherein The collision effect information includes collision valid information and collision invalid information. The generating of the collision effect information according to the maximum second-type collision evaluation value and the total collision evaluation value includes: Determine whether the total collision evaluation value is greater than a first preset evaluation threshold and whether the maximum second-type collision evaluation value is greater than a second preset evaluation threshold; If both are greater, determine that the collision effect information is collision valid information; If not greater, determine that the collision effect information is collision invalid information.

14. A robot, characterized in that, Includes: Collision part; A plurality of collision sensors, which are installed at intervals on the collision part; A signal conditioning circuit, electrically connected to each of the collision sensors; A controller, electrically connected to the signal conditioning circuit, and configured to execute the method for determining the collision position according to any one of claims 1 to 13.

15. A non-volatile readable storage medium, characterized in that, The non-volatile readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to execute the method for determining the collision position according to any one of claims 1-13.

Images