A collision detection system for evaluating obstacle recognition capabilities of intelligent mobile robots

Abstract

This invention discloses a collision detection system for evaluating the obstacle recognition ability of intelligent mobile robots, comprising: a base plate, guardrails, a base frame, simulated obstacles, a rigid structural cylinder, a force sensor, and a camera. Two parallel guardrails are arranged on the base plate to form a test area. The starting test point for the robot under test is set at the front of the test area. The base frame and the simulated obstacle are located at the rear of the test area, with an avoidance exit provided. A guide rail groove for mounting the force sensor is formed on the simulated obstacle. The rigid structural cylinder is fixedly connected in front of the force sensor as the collision contact object with the robot under test. The camera is positioned directly above the rigid structural cylinder and mounted on the base frame. This invention enables comprehensive testing of the robot's obstacle avoidance ability.

Description

Technical Field

[0001] This invention relates to a collision detection system for evaluating the obstacle recognition capabilities of intelligent mobile robots. Background Technology

[0002] In recent years, the market demand for mobile robots has continued to grow due to an aging population, rising labor costs, and increasing consumer demand for intelligent services. This demand is particularly acute in industries heavily reliant on human labor. Simultaneously, differences in needs across different fields and application scenarios are becoming increasingly apparent, with the market showing greater interest in innovative robot products with broad application potential. Currently, mobile robots have broad application prospects in various fields, including medical services, education services, catering services, hotel services, and retail services. With continuous technological advancements and the expansion of application scenarios, the market demand for mobile robots will continue to grow.

[0003] Mobile robots are often used in scenarios involving multiple types of obstacles. Their high degree of automation and intelligence depends on their ability to efficiently perceive their surroundings, quickly identify obstacles, and effectively avoid them. Currently, mobile robots primarily employ obstacle avoidance measures such as stopping in place after obstacle detection, replanning a route to bypass obstacles, and stopping and reversing after obstacle detection. Throughout these obstacle avoidance processes, it is crucial to ensure that none of these measures result in collisions between the mobile robot and its surroundings. Collision testing of obstacle avoidance capabilities can be conducted by building a simulated collision environment system to evaluate the mobile robot's performance.

[0004] Currently, most research and development in the industry on robot obstacle avoidance or collision avoidance function testing only provides tests with fixed obstacles and single variables. For example:

[0005] 1. A robot collision testing method and system disclosed in Publication No. CN117885137A involves the following steps: During the robot's movement towards an empty obstacle placement area, if a sensor sends first information indicating that the robot has reached a preset position, the system controls an obstacle driving device to activate and place an obstacle in the obstacle placement area. After the robot stops, the system obtains the distance between the robot's stopping position and the obstacle. The system determines the robot's collision avoidance capability at the current detection distance based on this distance, where the current detection distance is the distance between the sensor and the obstacle placement area. This method utilizes an obstacle driving device to temporarily or suddenly place obstacles in the obstacle placement area, replacing the possibility of a person suddenly appearing or carrying an object, thereby ensuring the safety of robot collision avoidance capability testing.

[0006] 2. CN116352756A discloses an indoor scene intelligent service robot obstacle avoidance function detection system and method. The detection system includes a power supply module, an image capture module, a force sensing module, a high-speed data acquisition module, a computer module, a fixed base, an obstacle substrate, and a fence. The fence encloses a collision scene testing activity fixture area with an exit. A simulated obstacle is installed at the exit of the fence on the fixed base. The obstacle substrate, connected to the force sensing module, is installed on the simulated obstacle. The force sensing module is connected to the high-speed data acquisition module. Both the high-speed data acquisition module and the image capture module are connected to the computer module. The power supply module supplies power to the image capture module, which monitors in real time whether the robot collides with the simulated obstacle.

[0007] 3. CN115824548A discloses a mobile robot collision safety testing and evaluation method and detection system. This method combines simulated collision and actual collision testing. It constructs a robot simulation model library, refines the simulation collision model using actual collision results, and conducts multi-mode simulated collision tests based on the optimized model. The robot collision force detection device used in actual collision testing includes a robot acceleration and angular velocity acquisition module, a collision force and torque acquisition module, a power supply module, and a measurement and control module. These modules are used to conduct actual collision tests and collect robot motion process parameters and collision parameters. The method also discloses a mobile robot collision safety detection system, including a cloud platform, an industrial control computer, and a robot collision force detection device, enabling full-process collision safety testing and evaluation of the robot. By combining actual and simulated collision testing, it enriches the types of robot collision tests and reduces the cost of robot collision testing.

[0008] Currently, most research and development in the industry on methods and devices for detecting obstacle avoidance or collision avoidance functions of mobile robots only considers situations where the mobile robot collides or comes to a complete stop, without taking into account the possible scraping or collision behaviors that may occur during the robot's detour or other avoidance actions. In addition, most devices are limited to the design of single obstacles and fixed obstacles, and do not take into account the situation of multiple obstacles of different sizes and shapes, making it difficult to test the robot's ability to recognize multiple obstacles and complete obstacle avoidance. Summary of the Invention

[0009] The purpose of this invention is to provide a collision detection system for evaluating the obstacle recognition ability of intelligent mobile robots. It can detect when a mobile robot stops, retreats, or collides with an obstacle, as well as the collision direction and magnitude of the collision force, thus achieving a comprehensive test of the robot's obstacle avoidance ability.

[0010] The technical solution of the present invention is as follows:

[0011] A collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot includes: a base plate, guardrails, a base frame, simulated obstacles, a rigid cylindrical structure, a force sensor, and a camera. The base plate has two parallel guardrails forming a rectangular test area for the robot under test. A starting test point for the robot is located at the front of the test area enclosed by the two guardrails. The base frame and the simulated obstacle are located at the rear of the test area. The base frame is constructed as a frame structure, creating exits on both sides of the simulated obstacle for the robot to exit the test area. The simulated obstacle has guide rail slots for mounting the force sensor. The rigid cylindrical structure is fixedly connected in front of the force sensor as a collision contact with the robot under test. A detachable clamping device can be used to connect the rigid cylindrical structure and the force sensor into a single force-bearing unit. The camera is positioned directly above the rigid cylindrical structure and mounted on the base frame.

[0012] The contact point between the rigid cylindrical structure and the force sensor is the force contact point of the sensor. When the robot under test collides with the rigid cylindrical structure, the force received by the force sensor at the force contact point is recorded as the test force F. meas The collision force exerted by the robot under test on the rigid cylindrical structure is denoted as the actual collision force F. coll The test force F meas The actual collision force F coll Let the included angle be θ, then we have

[0013] This invention includes guardrails to ensure the robot operates within a designated area during testing. Real-time monitoring via cameras records the robot's trajectory, allowing the system to determine whether the robot will stop, retreat, detour, or collide with an obstacle when encountering it. This information is used to evaluate the robot's obstacle avoidance capabilities. Upon collision, the system can determine the direction of the collision (head-on or side-on) and the magnitude of the impact force. When the robot detours around an obstacle, an exit at the rear of the testing area is provided to assess its detour ability.

[0014] To facilitate the rapid acquisition of the collision direction when the robot under test collides with a simulated obstacle, the upper surface of the top of the rigid structural cylinder of this invention has a 360° angle scale line. The center of this angle scale line is located on the central axis of the rigid structural cylinder. When the robot under test collides with the rigid structural cylinder, the collision contact point between the robot under test and the rigid structural cylinder is acquired by the camera, and the test force F is obtained. meas The actual collision force F coll The included angle θ is used to facilitate the calculation of the actual collision force value.

[0015] The force sensor of the present invention is preferably a force sensor with an S-shaped cantilever shear structure.

[0016] The simulated obstacles of the present invention include plate-shaped obstacles, cylindrical obstacles, and prismatic obstacles, which can simulate obstacles of different shapes, sizes, and types, such as walls, doors, tables, furniture, and pedestrians, for the robot under test to identify and detect different obstacles.

[0017] The force sensor of the present invention is height-adjustably installed in the guide rail groove, and its specific installation height is determined according to the height of the robot being tested, thus making it applicable to the detection of robots of different heights.

[0018] The rigid structural cylinder of the present invention has a height ≤150mm and a diameter ≤50mm. Specifically, it can be made of stainless steel metal parts. The size of the rigid structural cylinder should be much smaller than that of the robot under test to avoid the robot misidentifying obstacles. When the rigid structural cylinder is subjected to external force, it can maintain good force transmission, reduce the attenuation of force in the object, and improve the accuracy of collision force value testing.

[0019] The base frame of the present invention includes two uprights, on which a height-adjustable crossbar is installed. The camera is mounted on the crossbar, and the height-adjustable camera is designed to adapt to different types of robot testing.

[0020] The simulated obstacle of the present invention is fixedly connected to the base frame. Specifically, an adjustable or replaceable obstacle mounting bracket can be provided to adapt to different simulated obstacles.

[0021] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0022] 1. This invention comprehensively considers various movement modes of mobile robots. Using a single detection system, it can obtain various scenarios when the robot encounters an obstacle, such as stopping, retreating, circumventing, or colliding, to evaluate the robot's obstacle avoidance capability. In the event of a collision, the rigid cylindrical structure used in this invention can easily, quickly, and accurately determine whether the robot will collide head-on or sideways with the obstacle, and the magnitude of the collision force, especially accurately detecting the magnitude of the side collision force. When the robot circumvents an obstacle, its circumvention capability is tested at an obstacle avoidance exit at the rear of the test area.

[0023] 2. By replacing simulated obstacles, this invention can be used to simulate various obstacle avoidance scenarios. By using obstacles of different sizes, such as plate-shaped, cylindrical, and prismatic, it can simulate obstacles such as table legs and pedestrians, thereby more comprehensively testing the obstacle avoidance capabilities of mobile robots.

[0024] 3. The force sensor and camera height positions of the present invention are adjustable to adapt to testing mobile robots of different sizes and types.

[0025] 4. The present invention uses a rigid cylindrical structure as the collision contact object, which can maintain good force transmission, reduce the attenuation of force in the object, and improve the accuracy of collision force value testing. Attached Figure Description

[0026] Figure 1 This is a three-dimensional structural diagram of a collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot, provided by the present invention.

[0027] Figure 2 for Figure 1 A three-dimensional structural diagram from another angle;

[0028] Figure 3 The diagram shows the state of the robot under test as it detours around the exit.

[0029] Figure 4 A schematic diagram of the connection structure between a force sensor and a rigid cylindrical structure.

[0030] Figure 5 This is a force analysis diagram of the robot under test when it collides head-on with a rigid cylindrical structure.

[0031] Figure 6 This is a force analysis diagram of the robot under test when it collides sideways with a rigid cylindrical structure.

[0032] Explanation of reference numerals in the attached figures:

[0033] 1-Base plate; 2-Guardrail; 3-Base frame; 4-Simulated obstacle; 4a-Guide rail groove; 5-Avoidance exit; 6-Rigid structure cylinder; 7-Force sensor; 7a-Fixing clamping device; 8-Camera; 9-Robot under test. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0035] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0036] It should be understood that the invention can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0037] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising” and / or “including,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0038] To fully understand this invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by this invention. Optional embodiments of the invention are described in detail below; however, in addition to these detailed descriptions, the invention may have other embodiments.

[0039] like Figures 1 to 4 As shown, a collision detection system for evaluating the obstacle recognition ability of an intelligent mobile robot includes: a base plate 1, guardrails 2, a base frame 3, simulated obstacles 4, a rigid structural cylinder 6, a force sensor 7, and a camera 8. Two parallel guardrails 2 are set on the base plate 1 to form a rectangular test area for the robot under test 9. The starting test point of the robot under test 9 is set at the front of the test area enclosed by the two guardrails 2. The base frame 3 and the simulated obstacle 4 are set at the rear of the test area. The base frame 3 is constructed as a frame structure to form avoidance exits 5 for the robot under test 9 to exit the test area on both sides of the simulated obstacle 4. The simulated obstacle 4 has guide rail grooves 4a for mounting the force sensor 7. The rigid structural cylinder 6 is fixedly connected to the front of the force sensor 7 as a collision contact object with the robot under test 9. A detachable fixing clamping device 7a can be used to connect the rigid structural cylinder 6 and the force sensor 7 into a force-bearing whole. The camera 8 is set directly above the rigid structural cylinder 6 and is mounted on the base frame 3.

[0040] The contact point between the rigid cylinder 6 and the force sensor 7 is the force contact point of the sensor. When the robot under test 9 collides with the rigid cylinder 6, the force received by the force sensor 7 at the force contact point is recorded as the test force F. meas The collision force exerted by the robot on the rigid cylindrical structure 6 is denoted as the actual collision force F. coll The test force F meas The actual collision force F coll Let the included angle be θ, then we have

[0041] Figure 5 The diagram shows the forces acting on the tested robot 9 when it collides head-on with the rigid cylindrical structure 6. At this point, θ = 0, and the actual collision force F is... coll =Test force F meas .

[0042] Figure 6 The diagram shows the forces acting on the tested robot 9 during a lateral collision with the rigid cylindrical structure 6. The tangent point of the lateral collision is the point of contact between the tested robot 9 and the rigid cylindrical structure 6 at this moment. The actual collision force at this point is... Where F′ meas This refers to the force measured by force sensor 7 during a side collision.

[0043] To facilitate the rapid acquisition of the collision direction when the tested robot 9 collides with the simulated obstacle 4, the upper surface of the top of the rigid cylindrical structure 6 has a 360° angle scale line. The center of this angle scale line is located on the central axis of the rigid cylindrical structure 6. When the tested robot 9 collides with the rigid cylindrical structure 6, the camera 8 acquires the collision contact point between the tested robot 9 and the rigid cylindrical structure 6. The test force F can be obtained by the position of the collision contact point on the angle scale line at this time. meas The actual collision force F coll The included angle θ is used to facilitate the calculation of the actual collision force value.

[0044] In one embodiment, the force sensor 7 is preferably a force sensor with an S-shaped cantilever shear structure.

[0045] In some embodiments, the simulated obstacle 4 includes plate-shaped obstacles, cylindrical obstacles, and prismatic obstacles, which can simulate obstacles of different shapes, sizes, and types, such as walls, doors, tables, furniture, and pedestrians, for the robot under test to identify and detect different obstacles.

[0046] In one embodiment, the force sensor 7 is height-adjustably installed in the guide rail groove 4a. Its specific installation height is determined according to the height of the robot 9 being tested, thus making it applicable to the detection of robots of different heights. The connection method of the force sensor 7 in the guide rail groove 4a can be implemented using existing technology, which will not be elaborated here.

[0047] In some embodiments, the height of the rigid structural cylinder 6 is ≤150mm and the diameter is ≤50mm. Specifically, it can be made of stainless steel metal parts. The size of the rigid structural cylinder should be much smaller than that of the robot under test 9 to avoid the machine misidentifying obstacles. When the rigid structural cylinder 6 is subjected to external force, it can maintain good force transmission, reduce the attenuation of force in the object, and improve the accuracy of collision force value testing.

[0048] In one embodiment, the base frame 3 includes two uprights with height-adjustable crossbars mounted on them. A camera 8 is mounted on the crossbars, and the height-adjustable camera 8 is designed to adapt to different types of robot testing.

[0049] In one embodiment, the simulated obstacle 4 is fixedly connected to the base frame 3. Specifically, an adjustable or replaceable obstacle mounting bracket can be provided to adapt to different simulated obstacles 4.

[0050] The fixing method of the base plate and guardrail of the present invention, as well as the left-right distance and up-down height of the guardrail, can be set according to the size of the robot 9 being tested.

[0051] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot, characterized in that, include: The system comprises a base plate, guardrails, a base frame, simulated obstacles, a rigid cylindrical structure, a force sensor, and a camera. The base plate has two parallel guardrails forming a rectangular test area for the robot under test. The starting test point for the robot is located at the front of the test area enclosed by the two guardrails. The base frame and the simulated obstacles are located at the rear of the test area. The base frame is constructed as a frame structure, creating exits on both sides of the simulated obstacles for the robot to exit the test area. The simulated obstacles have guide rail slots for mounting the force sensor. The rigid cylindrical structure is fixedly connected in front of the force sensor as a collision contact point with the robot under test. The camera is positioned directly above the rigid cylindrical structure and mounted on the base frame. The contact point between the rigid cylindrical structure and the force sensor is the force contact point of the sensor. When the robot under test collides with the rigid cylindrical structure, the force received by the force contact point of the force sensor is recorded as the test force. The collision force exerted by the robot under test on the rigid cylindrical structure is recorded as the actual collision force. The test force With the actual collision force The included angle is denoted as ,but ; The upper surface of the rigid cylindrical structure has a 360° angle scale line, the center of which is located on the central axis of the rigid cylindrical structure. When the robot under test collides with the rigid cylindrical structure, the camera captures the point of contact between the robot under test and the rigid cylindrical structure, thus obtaining the test force. With the actual collision force The included angle .

2. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 1, characterized in that: The force sensor is an S-shaped cantilever shear structure force sensor.

3. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 1 or 2, characterized in that: The simulated obstacles include plate-shaped obstacles, cylindrical obstacles, and prismatic obstacles.

4. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 3, characterized in that: The force sensor is height-adjustably mounted within the guide rail groove.

5. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 1, characterized in that: The height of the rigid structural cylinder is ≤150mm and the diameter is ≤50mm.

6. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 1, characterized in that: The base frame includes two uprights, on which a height-adjustable crossbar is installed, and the camera is mounted on the crossbar.

7. The collision detection system for evaluating the obstacle recognition capability of an intelligent mobile robot according to claim 3, characterized in that: The simulated obstacle is fixedly connected to the base frame.

Images