Ultrasonic-based collision detection system and method for robot

Abstract

A method of detecting a collision on a robot housing is provided. The method includes: detecting an acoustic wave by a plurality of transducers mounted inside or outside a housing; estimating, by a processor, a collision location on the housing by time-of-arrival analysis of the acoustic waves detected by the plurality of transducers; determining, by a processor, a degree of collision based on the intensities of the acoustic waves detected by the plurality of transducers; and after the collision event is detected, the shell of the robot is automatically subjected to nondestructive testing, so that the damage degree of the collision event to the robot is evaluated.

Description

[0001] Cross - reference to related applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 423,502, filed on Nov. 08, 2022, which is incorporated herein by reference in its entirety. Technical field

[0003] This disclosure generally relates to the field of robotics. Specifically, this disclosure relates to a collision detection system and method for detecting collision events of a robot using ultrasonic detection technology. Background art

[0004] Robots are becoming increasingly indispensable in a wide variety of applications in many industries. In the working environment of robots, human error, component failure, control system failure, and other unpredictable events may cause the robot's trajectory to deviate or the workpiece to deflect, resulting in collisions. As industrial production evolves to include more human - robot collaboration, the uncertainties associated with human behavior in the robot's workspace exacerbate the need for a comprehensive safety protection system to protect both workers and robots.

[0005] Any failure to detect a collision and stop the robot's movement in a timely manner can cause serious damage to the robot system, valuable tools, and workpieces, and may pose a threat to the safety of human workers. Moreover, the process of assessing potential damage to robot components after a collision usually slows down the manufacturing speed of the entire production line. Therefore, there is an increasing demand for an automatic collision detection system in industrial robots: a system that can predict or immediately sense a collision, trigger protective measures, and estimate the degree of collision to help human operators determine whether repairs are needed.

[0006] Currently, there are two prominent solutions in the field of robot collision perception: contact - based methods and non - contact - based methods.

[0007] Non - contact - based methods can predict a collision before it occurs, enabling preventive safety measures to be taken to prevent the collision. These methods use lasers, infrared rays, or vision sensors to establish a "safety zone" around the workspace of the robotic arm. If an unknown object enters this "safety zone", the robotic arm will slow down or stop operating until the perceived danger is eliminated. This method offers significant safety advantages, but due to the use of sensor systems, it may reduce the efficiency of the robot and increase the overall system cost. This solution is mainly used in large automated workshops with a high degree of automation, but is less suitable for complex human - robot collaboration scenarios.

[0008] Contact methods detect collisions when they occur to trigger corresponding safety measures. Different from non-contact methods aimed at preventing collisions, contact methods strive to minimize the damage caused by the impact after a collision. These methods include current loop, electronic skin, and flexible joint types.

[0009] The current loop type transfers the collision torque through a spring to a set of speed reducers. When the torque reaches a certain level, the speed reducers start to rotate, dragging the motor to generate a current signal. Although this method has a low cost, its accuracy is insufficient and its load capacity is also small, making it only suitable for small robotic arms.

[0010] The electronic skin type involves installing a pressure sensor array on the surface of the robotic arm to detect external forces. Although this method has high sensitivity and accuracy, it has a complex system and high cost.

[0011] The flexible joint type installs torque sensors at the robot joints. When a collision is detected, the joints change from rigid to flexible, allowing the robot end to move freely along the external force, thereby reducing the damage caused by the collision. This method has a moderate cost and a wider application range than other types.

[0012] In view of this, there is a need in the art for a low-cost, efficient, and reliable solution for detecting robot collisions, which can be effectively implemented both in large automated workshops and in complex scenarios of human-robot collaboration. In addition, through the following detailed description of the present disclosure and the appended claims, and in conjunction with the drawings and the background art section, other desirable features and characteristics will become apparent. Summary of the Invention

[0013] This article provides a collision detection system and method that uses ultrasonic detection technology to detect collision events of a robot.

[0014] According to a first aspect of the present disclosure, a method for detecting a collision event on a robot housing is provided. The method includes: detecting sound waves through a plurality of transducers installed inside or outside the housing; estimating the collision position on the housing by a processor through time-of-arrival analysis of the sound waves detected by the plurality of transducers; determining the collision degree by the processor according to the intensity of the sound waves detected by the plurality of transducers; and automatically performing a non-destructive test on the robot housing after detecting a collision event to evaluate the damage degree caused by the collision event to the robot.

[0015] In one embodiment, the method further includes: when each of the plurality of transducers detects a sound wave, transmitting a response pulse from each of the plurality of transducers to the processor.

[0016] In one embodiment, the step of estimating the collision location on the housing by performing time-of-arrival analysis further includes: determining, by a processor, a time difference of response pulses received from each of the plurality of transducers relative to a collision event to calculate a distance from the collision location to each of the plurality of transducers; and locating the collision location on the housing based on the distances to each of the plurality of transducers.

[0017] In one embodiment, the response pulse has a signal amplitude defined by the intensity of the sound wave detected by the plurality of transducers; and wherein the processor is configured to collectively use the signal amplitude of the response pulse to calculate the degree of collision.

[0018] In one embodiment, the step of transmitting the response pulse to the processor further includes transmitting an electrical signal to the processor from time to time via wired or wireless communication, wherein the processor is configured to process the electrical signal to obtain the response pulse from each of the plurality of transducers.

[0019] In one embodiment, the plurality of transducers are spatially distributed on the housing to monitor the robot for detecting sound waves.

[0020] In one embodiment, the plurality of transducers include a plurality of ultrasonic transducers for detecting sound waves in the ultrasonic frequency range.

[0021] In one embodiment, the plurality of ultrasonic transducers are selected from the group consisting of piezoelectric transducers (PZT), piezoelectric polyvinylidene fluoride (PVDF) transducers, acoustic sensors, capacitive micromachined ultrasonic transducers (CMUT), and piezoelectric micromachined ultrasonic transducers (PMUT).

[0022] In one embodiment, the method further includes immediately pausing the operation of the robot when at least one of the plurality of transducers detects a sound wave to minimize damage to the robot and injury to personnel caused by the collision event.

[0023] In one embodiment, the step of performing a non-destructive test on the housing of the robot further includes: obtaining sound wave characteristics after the collision event via the plurality of transducers; and comparing, by the processor, the sound wave characteristics with previous data obtained before the collision event to evaluate the degree of damage caused to the robot by the collision event.

[0024] According to a second aspect of the present disclosure, a collision detection system for detecting collision events of a robot with a housing is disclosed. The collision detection system includes a plurality of transducers installed inside or outside the housing for detecting sound waves; and a processor configured to execute a method for determining whether a collision event has occurred on the housing, wherein the method includes the following steps: performing time-of-arrival analysis on the detected sound waves to determine the collision location; performing signal analysis on the detected sound waves to determine the degree of collision; and automatically performing a non-destructive test on the housing of the robot after detecting a collision event to evaluate the degree of damage caused to the robot by the collision event.

[0025] This "Summary of the Invention" section is provided to introduce some concepts in a simplified form, which will be further described in the "Detailed Description" section below. This "Summary of the Invention" section is not intended to identify the key features or essential features of the subject matter claimed in this application, nor is it intended to be used as an aid in determining the scope of the subject matter claimed in this application. Other aspects and advantages of the present invention are disclosed as shown in the embodiments hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings include figures for further illustrating and clarifying the above and other aspects, advantages, and features of the present disclosure. It is understood that these figures only depict some embodiments of the present disclosure and are not intended to limit its scope. It is also understood that these figures are shown for simplicity and clarity and are not necessarily drawn to scale. The present disclosure will now be described and explained with additional specificity and detail by using the drawings, wherein:

[0027] Figure 1 is a robot with a collision detection system according to some embodiments of the present disclosure.

[0028] Figure 2 is a photograph of the housing surrounding the robot according to some embodiments of the present disclosure.

[0029] Figure 3 is an internal view of the robot with a plurality of transducers for collision detection Figure 2 of

[0030] Figure 4 is a graph of the collision location error. DETAILED DESCRIPTION

[0031] The following detailed description is merely exemplary in nature and is not intended to limit the present disclosure or its application and / or uses. It is understood that there are numerous variations. This detailed description enables a person of ordinary skill in the art to implement the exemplary embodiments of the present disclosure without undue experimentation, and it is understood that various changes or modifications can be made to the functions and structures described in the exemplary embodiments without departing from the scope of the present disclosure as set forth in the appended claims.

[0032] Benefits, advantages, solutions to problems, and any element or elements that may cause any benefit, advantage, or solution to occur or become more apparent should not be construed as critical, essential, or core features or elements of any or all of the claims. The present invention is defined solely by the appended claims, including any amendments made during the pendency of this application and all equivalents of those claims as granted.

[0033] As used herein, the term "processor" generally refers to all types of digital processing devices, including but not limited to: microcontroller units, custom integrated circuits, digital signal processors, field programmable gate arrays, application specific integrated circuits, central processing units, graphics processing units, computer devices, programmable I / O devices, other semiconductor devices, or any combination thereof.

[0034] In the context of describing the present invention (especially in the context of the following claims), the use of the terms "a", "an", "the", and "at least one" and similar references will be construed to cover both the singular and plural cases unless otherwise stated herein or clearly contradicted in the context. The terms "comprising", "having", and "including", or any other variant thereof, will be construed as open terms (i.e., meaning "including but not limited to") unless otherwise specified. The use of any and all examples or exemplary language (e.g., "such as") provided herein is merely intended to better illustrate the present invention and does not limit the scope of the present invention unless otherwise stated. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the present invention.

[0035] All terms (including technical and scientific terms) used in the embodiments of the present invention have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention pertains, unless otherwise defined.

[0036] Based on the foregoing background, there is a desire for a collision detection system for a robot. Specifically, the collision detection system is a low-cost, efficient, and reliable system for detecting robot collisions. It can be effectively implemented both in large automated workshops and in complex scenarios involving human-robot collaboration.

[0037] The present disclosure relates to a collision detection system and a method capable of detecting robot collision events. Figure 1 A robot 50 having a collision detection system according to an embodiment of the present invention is shown. The robot 50 includes one or more movable elements 51, immovable elements 52, and a housing 53 that at least partially surrounds the one or more movable elements 51. Figure 2An exemplary housing 53 surrounding the robot 50 is shown. In some embodiments, the housing 53 may also surround the immovable element 52. As Figure 1 shown in the illustrated embodiment, the robot 50 is a robotic arm for controlling the physical position and orientation of the head 54. The head 54 may include functional devices capable of performing one or more tasks.

[0038] Generally speaking, the robot 50 is relatively complex and has various mechanical and electronic devices inside. The available space for installing sensors is limited, so sensors that are small enough for the robot 50 and can be conveniently installed without occupying too much space and effort are needed.

[0039] In some embodiments, the collision detection system includes a plurality of transducers 100 installed inside or outside the housing 53 for detecting sound waves. The term "sound wave" does not denote waves of a specific frequency, and preferably, the sound waves are ultrasonic waves within the ultrasonic frequency range. Thus, the present disclosure is fundamentally supported by ultrasonic detection technology. When a collision event occurs, the sound waves generated by the collision event will propagate through the housing 53. For example, when the one or more movable elements 51 collide with a person, the contact on the housing 53 will generate sound waves in all directions. Sound waves, especially ultrasonic waves, propagate very quickly in solid materials (such as metals or composite materials) commonly used as the housing 53 of the robot 50. Therefore, after a collision occurs, the plurality of transducers 100 can detect the collision within a few milliseconds. For example, the wave speed in an aluminum plate is about 5 km / s, and the response time can be about 0.02 ms.

[0040] Since it is assumed that the sound waves can be ultrasonic waves, the plurality of transducers 100 may include a plurality of ultrasonic transducers for detecting sound waves within the ultrasonic frequency range. In some embodiments, the plurality of ultrasonic transducers are selected from the group consisting of piezoelectric transducers (PZT), piezoelectric polyvinylidene fluoride (PVDF) transducers, acoustic sensors, capacitive micromachined ultrasonic transducers (CMUT), piezoelectric micromachined ultrasonic transducers (PMUT), and other types of ultrasonic transducers. For those traditional collision sensors, the cost is relatively high. For example, the most common flexible joints usually cost thousands of dollars and cannot be widely used in the industry. According to the present disclosure, the installation cost of using ultrasonic transducers is low, and ultrasonic sensors are easily available in the market. Each individual transducer costs less than $1.

[0041] The plurality of transducers 100 need to be in contact with the housing 53, but there is no limitation as to whether the plurality of transducers 100 are mounted inside or outside the housing 53. The positions of the plurality of transducers 100 are determined to cover the entire surface of the housing 53 so as to easily estimate the collision position. Preferably, the plurality of transducers 100 are evenly mounted inside or outside the housing 53 and are spatially distributed on the housing 53 in a predetermined pattern for the robot 50 to detect sound waves. Figure 3 An internal view of the robot 50 is shown, which has a plurality of transducers 100 mounted inside the housing 53 for collision detection. This example provides 8 ultrasonic transducers mounted on a 30 cm long aluminum curved housing 53.

[0042] The collision detection system further includes a processor 120 configured to execute a method for determining whether a collision event has occurred on the housing 53. When any one of the plurality of transducers 100 detects a sound wave, the transducer will send a response pulse to the processor 120. In some embodiments, the plurality of transducers 100 are configured to send electrical signals to the processor 120 from time to time via wired or wireless communication. For the case of wireless communication, the plurality of transducers 100 are electrically connected to one or more wireless transmitters, such as Bluetooth transceivers, Wi-Fi transceivers, Zigbee transceivers, and / or other similar types of wireless transceivers configured to communicate on a wireless network.

[0043] The processor 120 is configured to process the electrical signals to obtain response pulses from each of the plurality of transducers 100. The present disclosure advantageously provides that the processor 120 evaluates the response pulses from the plurality of transducers 100 to determine the collision position and the degree of collision. In this preferred embodiment, the processor 120 is configured to perform time-of-arrival analysis on the detected sound waves to determine the collision position; and perform signal analysis on the detected sound waves to determine the degree of collision.

[0044] For the collision position, the distances between the collision point and each of the plurality of transducers 100 are different. When a collision occurs, the times required for the sound waves to reach each of the plurality of transducers 100 are different. By calculating the time differences of the detected sound waves between each of the plurality of transducers 100, the collision position can be located based on the time-of-arrival analysis. Therefore, the processor is configured to determine the time differences of the response pulses received by each of the plurality of transducers 100 with respect to the collision event to calculate the distances from the collision position to each of the plurality of transducers 100. Based on the calculated distances, the collision position can be located. Figure 4 A graph showing the collision localization error is provided, all of which are less than 5 mm.

[0045] For the degree of collision, different collision forces or torques will generate different acoustic wave intensities. By performing signal analysis on the acoustic waves detected by the plurality of transducers 100, the degree of collision can be evaluated. When the collision force is strong, the acoustic wave has a large intensity and can be detected by the plurality of transducers 100 accordingly. Based on the detected acoustic wave, the plurality of transducers 100 emit response pulses with different signal amplitudes. Therefore, the response pulse has a signal amplitude defined by the intensity of the acoustic wave detected by the plurality of transducers 100, and this response amplitude is transmitted to the processor 120 to calculate the degree of collision. The signal amplitudes from the plurality of transducers 100 are collectively evaluated, and the corresponding distances from each to the collision location are also considered to determine the degree of collision.

[0046] Another aspect of the present disclosure provides non-destructive testing after a collision event is detected. The processor 120 is configured to automatically perform non-destructive testing on the housing 53 of the robot 50 to evaluate the degree of damage caused to the robot 50 by the collision event. If there is damage to the structure, this can be detected by analyzing the acoustic waves detected by each of the plurality of transducers 100. Through non-destructive testing, the robot 50 can obtain preliminary test results regarding the degree of collision damage before a human operator arrives at the scene. Specifically, the non-destructive testing of the housing 53 includes obtaining the acoustic wave characteristics after the collision event through the plurality of transducers 100; and comparing the acoustic wave characteristics with previous data obtained before the collision event through the processor 120 to evaluate the degree of damage caused to the robot 50 by the collision event. The non-destructive testing can be completed within a few minutes. If the robot 50 is not damaged, it can resume work immediately. If it is determined that the degree of damage is high, the processor 120 is configured to notify a human operator to come and evaluate the degree of damage and what measures need to be taken to repair the robot 50.

[0047] In some embodiments, the processor 120 is configured to immediately pause the operation of the robot 50 when at least one of the plurality of transducers 100 detects an acoustic wave, in order to minimize the damage caused to the robot 50 by the collision event and the harm caused to individuals. Since ultrasonic waves propagate very fast, any one of the plurality of transducers 100 can detect a collision within a few milliseconds. The mechanism of immediately pausing the robot 50 can protect the robot 50 from further damage. If the collision involves an individual, the pause can also minimize the harm caused.

[0048] In the illustrated embodiment, the robot 50 is a robotic arm and the testing is limited to some actions. However, obviously, the collision detection system can be applied to other larger robot systems and humanoid robots without departing from the scope and spirit of the present disclosure. Therefore, the collision detection system can achieve a full-body collision sensing ability.

[0049] This document describes a collision detection system and method for detecting collision events of a robot 50 using ultrasonic detection technology according to the present disclosure. It will be apparent that the above-described variations and other features and functions or their alternatives can be integrated into a humanoid robot or other automated systems. Therefore, this embodiment should be considered illustrative in all respects and not restrictive. The scope of the present disclosure is indicated by the appended claims rather than the foregoing description, and thus, all variations falling within the equivalent meaning and scope of the claims are intended to be covered therein.

Claims

1. A method for detecting a collision event on the housing of a robot, comprising: detecting sound waves by a plurality of transducers installed inside or outside the housing; estimating a collision location on the housing by a processor through time-of-arrival analysis of the sound waves detected by the plurality of transducers; determining a collision severity by the processor based on the intensity of the sound waves detected by the plurality of transducers; and automatically performing a non-destructive test on the housing of the robot after detecting a collision event to evaluate the damage degree caused by the collision event to the robot.

2. The method according to claim 1, further comprising the step of transmitting a response pulse from each of the plurality of transducers to the processor when the sound wave is detected by each of the plurality of transducers.

3. The method according to claim 2, wherein, the step of estimating a collision location on the housing by performing time-of-arrival analysis further comprises: determining, by the processor, a time difference of receiving the response pulse from each of the plurality of transducers relative to the collision event to calculate a distance from the collision location to each of the plurality of transducers; and determining a collision location on the housing based on the distances to each of the plurality of transducers.

4. The method according to claim 2, wherein, the response pulse has a signal amplitude defined by the intensity of the sound wave detected by the plurality of transducers; and wherein the processor is configured to collectively use the signal amplitude of the response pulse to calculate the collision severity.

5. The method according to claim 2, wherein, the step of transmitting the response pulse to the processor further comprises transmitting an electrical signal to the processor from time to time through wired or wireless communication, wherein the processor is configured to process the electrical signal to obtain the response pulse from each of the plurality of transducers.

6. The method according to claim 1, wherein, the plurality of transducers are spatially distributed on the housing to monitor the robot for detecting sound waves.

7. The method according to claim 6, wherein, the plurality of transducers include a plurality of ultrasonic transducers for detecting sound waves in the ultrasonic frequency range.

8. The method according to claim 7, wherein, the plurality of ultrasonic transducers are selected from the group consisting of piezoelectric transducers (PZT), piezoelectric polyvinylidene fluoride (PVDF) transducers, acoustic sensors, capacitive micromachined ultrasonic transducers (CMUT), and piezoelectric micromachined ultrasonic transducers (PMUT).

9. The method according to claim 1, further comprising immediately pausing the operation of the robot when at least one of the plurality of transducers detects the sound wave to minimize damage to the robot and injury to personnel caused by the collision event.

10. The method according to claim 1, wherein, the step of performing a non-destructive test on the housing of the robot further comprises: acquiring sound wave characteristics after the collision event by the plurality of transducers; and The processor compares the acoustic wave characteristics with previous data acquired before the collision event to evaluate the degree of damage to the robot caused by the collision event.

11. A collision detection system for detecting a collision event of a robot having a housing, comprising: a plurality of transducers for detecting acoustic waves, mounted inside or outside the housing; and a processor configured to execute a method for determining whether a collision event occurs on the housing, wherein the method comprises the following steps: performing time-of-arrival analysis on the detected acoustic waves to determine the collision location; performing signal analysis on the detected acoustic waves to determine the degree of collision; and after detecting a collision event, automatically performing non-destructive testing on the housing of the robot to evaluate the degree of damage to the robot caused by the collision event.

12. The collision detection system according to claim 11, wherein each transducer among the plurality of transducers sends a response pulse to the processor when detecting the acoustic wave.

13. The collision detection system according to claim 12, wherein the step of performing time-of-arrival analysis further comprises: determining the time difference of receiving the response pulse from each transducer among the plurality of transducers relative to the collision event to calculate the distance from the collision location to each transducer among the plurality of transducers; and locating the collision location on the housing based on the distances to each transducer among the plurality of transducers.

14. The collision detection system according to claim 12, wherein the response pulse has a signal amplitude defined by the intensity of the acoustic wave detected by the plurality of transducers; and wherein the processor is configured to collectively use the signal amplitude of the response pulse to calculate the degree of collision.

15. The collision detection system according to claim 12, wherein the plurality of transducers are configured to transmit electrical signals to the processor from time to time through wired or wireless communication; and wherein the processor is configured to process the electrical signals to obtain response pulses from each transducer among the plurality of transducers.

16. The collision detection system according to claim 11, wherein the plurality of transducers are spatially distributed on the housing to monitor the robot for detecting acoustic waves.

17. The collision detection system according to claim 16, wherein the plurality of transducers include a plurality of ultrasonic transducers for detecting acoustic waves in the ultrasonic frequency range.

18. The collision detection system according to claim 17, wherein the plurality of ultrasonic transducers are selected from the group consisting of piezoelectric transducers (PZT), piezoelectric polyvinylidene fluoride (PVDF) transducers, acoustic sensors, capacitive micromachined ultrasonic transducers (CMUT), and piezoelectric micromachined ultrasonic transducers (PMUT).

19. The collision detection system according to claim 1, wherein The processor is configured to immediately pause the operation of the robot when at least one of the plurality of transducers detects an acoustic wave, so as to minimize damage to the robot and injury to individuals caused by the collision event.

20. The collision detection system according to claim 1, wherein, the step of performing a non-destructive test on the housing of the robot further includes: acquiring acoustic wave characteristics after the collision event through the plurality of transducers; and comparing the acoustic wave characteristics with previous data acquired before the collision event by the processor to evaluate the degree of damage caused to the robot by the collision event.

Images