Height change detection system, robot including same, and associated method

Abstract

The present disclosure provides a height change detection system, a robot including the same, and an associated method. A height change detection system for a robot having an operating system is provided. The height change detection system includes a sensor, a processor electrically connected to the sensor, and a memory. The memory has instructions that, when executed by the processor, cause the processor to perform operations including: detecting an edge of a structure with the sensor; classifying a height change associated with the edge; and adjusting the operating system based on the height change.

Description

Technical Field

[0001] The systems, devices, and methods disclosed herein assist, at least in part, in allowing robots (whether they are bipedal robots, wheeled robots, or quadrupedal robots) to traverse structures. Background Art

[0002] Most robots lack the ability to handle the various challenges of human-centric environments, such as sensing sudden changes in altitude. Humans do well in this area due to extensive experience in both indoor and outdoor environments that are designed to be human-friendly. However, these are not obvious to robots unless complex algorithms are employed and combined with a set of expensive sensors to perform these tasks. During autonomous navigation in and around buildings, robots will need to detect and model changes in altitude. These changes in altitude can include ascending and descending stairs, curbs, and cliffs. Knowing the location of stairs, curbs, and cliffs can ensure that the robot moves efficiently and also allow for traversal of difficult environments. Existing algorithms are designed for either ascending situations or descending situations, but not both. In addition, existing algorithms may require sensors to be positioned outside the robot, which may require changes to existing infrastructure.

[0003] For example, a sensor may be placed directly above a set of stairs to obtain the properties of the stairs. In order for the robot to receive this information, there may be communication between the infrastructure and the robot. However, this is not very useful when no such infrastructure exists. In this case, the robot may not be able to operate effectively in these environments. Therefore, understanding the environment from the robot's vantage point may be beneficial for the robot to operate independently and autonomously.

[0004] It is with respect to these and other considerations that the instant disclosure herein is made. Summary of the invention

[0005] Any suitable robot can be used herein. The robot can be partially or fully autonomous. More specifically, the disclosed concept provides a height change detection system for a robot with an operating system. The height change detection system includes a sensor, a processor, and a memory, the memory enabling the processor to detect the edge of a structure using the sensor, classify the height change associated with the edge (e.g., up or down stairs, curbs, or cliffs), and adjust the operating system based on the height change. The sensor can be a camera, sonar, radar, or lidar sensor, or the system can include both a camera and a lidar sensor, which together detect the edge of the structure. Any suitable sensor or combination of sensors can be used herein. In this way, data from each of the camera and lidar sensor is fused by the processor, thereby allowing the edge of the structure to be detected in a relatively robust manner. Therefore, the disclosed method can extract high-fidelity height properties using the fusion of point cloud data from multiple sensors.

[0006] Thus, the disclosed systems, robots including the systems, and associated methods can rely on edge-based detection to traverse structures (e.g., up and down stairs, curbs, and cliffs), whereas typical systems for robots rely on plane-based detection. According to the disclosed concepts, edge-based detection allows the robot to traverse structures with significantly less required computation than typical systems.

[0007] Furthermore, in one example, adjustments to the robot's operating system, which can be configured to control a pair of legs of the robot, are advantageously made in real time. That is, when the sensor detects an edge, the processor causes the operating system to make adjustments in real time to traverse the structure. The robot's processor can also be configured to determine an up-and-down angle associated with the structure, and can then cause the robot to traverse the structure at the up-and-down angles. Thus, it should be appreciated that the robot is relatively less likely to tip over than a typical robot.

[0008] Additionally, in one example, a sensor suite including at least one of a camera and a lidar sensor is configured to be positioned at an angle between 35 and 55 degrees relative to the body of the robot, thereby allowing the sensor suite to scan a relatively large portion of the structure to be traversed. Furthermore, the processor of the system can be configured to adjust the tilt angle of either or both of the camera and lidar sensor in response to detecting an edge. Thus, the sensor suite is advantageously able to view more of the structure to be traversed as the robot ascends or descends, compared to typical systems that may not provide an adjustable sensor suite.

[0009] In another example, the robot's sensor suite is configured to scan multiple edges, including vertical and horizontal edges such as those associated with stairs, curbs, and cliffs. Additionally, the sensor suite is configured to continuously scan the structure as the robot traverses these structures, thereby continuously feeding data to the processor, thereby allowing the operating system to move, for example, the robot's legs in real time. It will be appreciated that such a system is relatively efficient and causes the robot to tip over far less often than a typical system.

[0010] These and other advantages of the present disclosure are provided in greater detail herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Detailed description is set forth with reference to the accompanying drawings. The use of the same reference numerals may indicate similar or identical items. Various embodiments may utilize elements and / or components other than those shown in the accompanying drawings, and some elements and / or components may not be present in various embodiments. The elements and / or components in the accompanying drawings are not necessarily drawn to scale. Throughout this disclosure, singular and plural terms may be used interchangeably, depending on the context.

[0012] Figure 1 is an isometric view of a robot and height change detection system thereof according to one non-limiting embodiment of the disclosed concept, the robot and the height change detection system being shown preparing to ascend a flight of stairs.

[0013] Figure 2 yes Figure 1 A simplified view of the robot and its height change detection system.

[0014] Figure 3 is another isometric view of the robot and its height change detection system, and is shown holding a package and ascending the flight of stairs.

[0015] Figure 4A is another isometric view of the robot and its height change detection system, and is shown as the robot prepares to descend a flight of stairs.

[0016] Figure 4B Is Figure 4A A 3D graphical representation of the data points captured by the robot’s sensor suite.

[0017] Figure 5 is another isometric view of the robot and its height change detection system, and is shown as the robot ascends a curb.

[0018] Figure 6 is another isometric view of the robot and its height change detection system, and is shown with the robot positioned next to a cliff.

[0019] Figure 7 An example process is shown in which altitude is classified by the altitude change detection system of the disclosed concept.

[0020] Figure 8 An example process is shown in which edges are detected by the height change detection system of the disclosed concept.

[0021] Fig. 9 An example algorithm that may be employed by a processor of a height change detection system of the disclosed concept is shown.

[0022] Fig.10 The process of fusing data from the camera and lidar sensor is shown.

[0023] Fig.11 Shows the chain that a height label goes through before being used to notify a lower-level control.

[0024] Fig.12 Another robot and height change detection system according to another non-limiting embodiment of the disclosed concept is shown.

[0025] Fig.13A and Fig. 13B Another robot and height change detection system according to another non-limiting embodiment of the disclosed concept is shown. DETAILED DESCRIPTION

[0026] As used herein, the term "coupled" shall mean connected together directly or via one or more intermediate parts or components. Depending on the context, "coupled" may also refer to a mechanical connection or an electrical connection (including both wired or wireless).

[0027] Figure 1 and Figure 2 An isometric and simplified view of a robot 2 and its height change detection system 4 at the base of a flight of stairs 100 are shown, respectively, in accordance with one non-limiting embodiment of the disclosed concept. Figure 2As shown, the robot 2 and its height change detection system 4 include a sensor suite 10, a processor 12, and a memory 14. In one example, the sensor suite 10 includes a camera (e.g., a stereo depth camera 20) and a lidar sensor 22, each electrically connected to the processor 12. The position and orientation of the camera 20 and the lidar sensor 22 can be used as parameters that can be tuned to obtain optimal detection and classification performance of the algorithm employed by the processor 12. In addition, the robot 2 may also include an operating system 30 electrically connected to each of the camera 20 and the lidar sensor 22, a pair of legs 32, 33 and a pair of arms 38, 39, each configured to be moved by the operating system 30, and a body 36 coupled to each of the pair of legs 32, 33 and the pair of arms 38, 39. In one example, the stereo depth camera 20 and the lidar sensor 22 are each coupled to the top of the body 36, and each is configured to scan the height change of the structure.

[0028] In addition, the memory 14 may include instructions that, when executed by the processor 12, cause the processor 12 to perform operations including detecting edges (e.g., horizontal edges 102, 106 and / or vertical edges 104, 108) of a structure (e.g., a flight of stairs 100) using at least one of the camera 20 and the lidar sensor 22, classifying height changes associated with the edges 102, 104, 106, 108, and adjusting the operating system 30 based on the height changes. When the edges 102, 104, 106, 108 (e.g., and all other edges) of the individual stairs of the flight of stairs 100 are detected, the processor 12 may extrapolate the geometry of each of the individual stairs. The algorithm employed by the processor 12 may be configured to detect which of the edges 102, 104, 106, 108 are horizontal edges and which are vertical edges in order to classify whether the robot 2 is going up or down. In addition, in a preferred example, the memory 14 also includes instructions that, when executed by the processor 12, cause the processor to detect the edges 102, 104, 106, 108 of the flight of stairs 100 using the camera 20, and to fuse data from the lidar sensor 22 with data from the camera 20 in order to classify height changes associated with the edges 102, 104, 106, 108.

[0029] Thus, the disclosed method can extract high-fidelity height properties using a fusion of point cloud data from multiple sensors (e.g., camera 20 and lidar sensor 22). The algorithms employed with the processor 12 of the disclosed system 4 are configured to be modular so that they can handle the absence of either sensor 20, 22 or the presence of both sensors 20, 22 (e.g., and also additional sensors not shown) to increase the ability of the robot 2 to traverse the flight of stairs 100 without tipping over. Thus, it should be understood that the system 4 can provide multiple levels of data to allow the operating system 30 to be appropriately adjusted and, therefore, beneficially enable the robot 2 to traverse the flight of stairs 100, whether the flight of stairs is inside or outside of a building.

[0030] The disclosed edge-based detection system is advantageous over typical systems (not shown) that typically rely on plane-based systems, in which the robot's sensors must see the entire face of the step because they are looking for a plane. In such systems (not shown), the robot cannot determine the exact dimensions of the staircase because the robot is looking for a plane. In contrast, in accordance with the disclosed concepts, edge-based detection requires fewer computations by the processor 12, thereby allowing the robot 2 to prepare to go up or down in real time, and also to move up or down faster. This corresponds to the robot 2 being able to operate in an environment that is generally more suitable for humans. In addition, the relatively sharp edges of the section of stairs 100 are generally consistent across all stairs (e.g., other stairs have similar properties, such as a face that may be generally perpendicular to the face extending upward). Therefore, there is no need to learn specific properties of the section of stairs 100. In addition, in the event that the robot 2 descends (e.g., Figure 4A ), edge-based detection may be more robust since the stairs may be more blurred in this case.

[0031] Thus, when the edge 102, 104, 106, 108 of the flight of stairs 100 is detected, adjustments to the operating system 30 may be performed in real time. Furthermore, adjustments to the operating system 30 are preferably performed prior to causing the pair of legs 32, 33 to traverse the flight of stairs 100. In contrast, a typical robot (not shown) is not typically so configured, and typically results in the robot not being able to detect sudden changes in height. As a result, the robot 2 is significantly less likely to tip over when ascending or descending than a typical system that relies on plane-based detection (e.g., detecting the entire plane of the stairs to indicate the robot's movement).

[0032] like Figure 3As shown, the robot 2 has begun to traverse (e.g., ascend) the flight of stairs 100 (e.g., one of the legs 32, 33 is on the first stair of the flight of stairs 100). This can be done in response to the camera 20 and / or the lidar sensor 22 scanning the flight of stairs 100 and detecting the edges 102, 104, 106, 108. As the robot 2 steps on each stair in the flight of stairs 100, the processor 12 of the robot 2 can anticipate the height and step depth of each stair, thereby allowing it to appropriately control the joint angles of the legs 32, 33 so that it can apply the appropriate torque and step on each of the stairs 100. In addition, the camera 20 and the lidar sensor 22 are configured to scan a relatively large portion of the flight of stairs 100. For example, the sensor suite 10 including the camera 20 and the lidar sensor 22 is preferably mounted to the body 36 of the robot 2 at an angle between 35 degrees and 55 degrees relative to the body 36. In one example, the sensor suite 10 is mounted to the body 36 at an angle of 45 degrees relative to the body 36. Therefore, the camera 20 and lidar sensor 22 are advantageously configured to scan a relatively large portion of the flight of stairs 100 as the robot 2 ascends.

[0033] Additionally, in one example, the memory 14 also includes instructions that, when executed by the processor 12, cause the processor 12 to detect other edges 104, 106, 108 (e.g., and all edges of the flight of stairs 100), and to continuously adjust the operating system 30 in real time based on height changes associated with the other edges 104, 106, 108. Thus, the robot 2 will be able to continuously monitor its ascent (e.g., and also descent, as will be referenced) and the height changes associated with the other edges 104, 106, 108. Figure 4A 108). In addition, the memory 14 may also include instructions that, when executed by the processor 12, cause the processor to adjust the tilt angle of either or both of the camera 20 and the lidar sensor 22 in response to detecting the edges 102, 104, 106, 108. Thus, as the robot 2 ascends the flight of stairs 100, the sensor suite 10 is advantageously able to observe more of the stairs 100, whereas many systems (not shown) typically have sensor suites that are maintained at a fixed angle relative to the body of the robot, which typically causes the robot to tip over because their field of view becomes relatively limited.

[0034] Therefore, in Figure 1 and Figure 3, the height change corresponds to the robot 2 ascending the flight of stairs 100. It will be appreciated that the processor 12 of the height change detection system 4 may be configured to save data for each of the stairs 100. Thus, the robot 2 may be configured to determine that there are a predetermined number of stairs (e.g., five stairs), and then once the robot 2 steps on the first stair, it will know how to traverse the remaining stairs. Furthermore, the robot 2 may be configured such that after it climbs the flight of stairs 100, the stored data may be used to allow it to climb the flight of stairs 100 again in a relatively efficient manner in the future. In other words, the edge-based height change detection system 4 learns from each successive encounter with the flight of stairs 100, as well as other height changes associated with other structures, as will be discussed below.

[0035] In another example embodiment, the disclosed edge-based height change detection system 4 for the robot 2 is also configured to be associated with other height changes. For example, Figure 4A The robot 2 is shown preparing to descend a flight of stairs 200 having multiple edges (three edges 202, 204, 206 are labeled). Figure 5 The robot 2 is shown ascending a curb 300 having at least one edge (eg, a vertical edge 302 and a horizontal edge 304), and Figure 6 The robot 2 is shown positioned next to a cliff 400 having an edge 402. Thus, the height change classified by the height change detection system 4 of the disclosed concept may be one of the number of ascending stairs 100, the number of descending stairs 200, the curb 300, and the cliff 400, wherein adjustments to the operating system 30 in each of these situations may be performed in real time when the edges 102, 104, 106, 108, 202, 204, 206, 302, 304, 402 of the corresponding structure 100, 200, 300, 400 are detected.

[0036] Figure 4B Is Figure 4A3D graphical representation of data points captured by the sensor suite 10 of the robot 2. As shown, the data points corresponding to the edges 202, 204, 206 are relatively clearly defined, thereby providing the robot 2 with an accurate picture of the geometry of each of the stairs. In addition, these data points are relatively close to the robot 2, thereby allowing the height change detection system 4 to accurately provide the robot's initial steps downward. In addition, as the robot 2 descends, the tilt angle of the sensor suite 10 is configured to change to correspond to the descent angle, so that the robot 2 will continuously detect the edges of the downwardly disposed stairs of the flight of stairs 200 to a relatively accurate degree (e.g., compared to a plane-based system). That is, as the robot 2 descends the flight of stairs 200, the sensor suite 10 will continuously scan the stairs 200 (e.g., and receive data corresponding to each individual stair toward which the robot 2 moves), and the processor 12 will adjust the operating system 30 in real time, thereby allowing the robot 2 to descend in a relatively efficient and reliable manner.

[0037] In addition, the memory 14 preferably also includes instructions that, when executed by the processor 12, cause the processor 12 to determine an ascending angle (e.g., relative to the flight of stairs 100 or the curb 300) or a descending angle (e.g., relative to the flight of stairs 200 or the cliff 400) corresponding to the change in height, and cause the pair of legs 32, 33 to traverse the structure 100, 200, 300, 400 at the ascending angle or the descending angle. That is, once the sensor suite 10 has scanned the structure 100, 200, 300, 400, the processor 12 determines the ascending or descending angle in real time and adjusts the operating system 30, which in turn causes the pair of legs 32, 33 to ascend or descend in a relatively efficient and reliable manner. In addition, relative to the cliff 400, if the processor 12 determines that the edge 402 is located at least a predetermined height above the ground, the processor 12 can determine that traversing the cliff 400 will cause the robot 2 to tip over, resulting in a situation where the robot 2 will decide not to traverse the cliff 400 (e.g., the robot 2 will move away from the edge 402).

[0038] In a preferred embodiment of the disclosed concept, the robot 2 is configured to adjust as it traverses each of the structures 100, 200, 300, 400. More specifically, the robot 2 has a center of gravity, and the step of traversing the pair of legs 32, 33 through the structures 100, 200, 300, 400 can be performed while the center of gravity of the robot 2 remains fixed relative to the pair of legs 32, 33. For example, when one of the legs 32, 33 begins to ascend, e.g. Figure 1In the example of stepping up the first stair of the stairs 100, one of the arms 38, 39 can move forward while the other of the arms 38, 39 moves backward, thereby allowing the center of gravity of the robot 2 to remain in a fixed position relative to the pair of legs 32, 33 and thereby minimizing the possibility of the robot 2 tipping over.

[0039] In another example, Figure 3 As shown, the pair of arms 38, 39 can hold an item, such as a package 150, as the robot 2 ascends the flight of stairs 100. According to the disclosed concepts, the pair of arms 38, 39 can extend a predetermined distance from the body 36 of the robot 2, and the predetermined distance can change as the robot 2 ascends the flight of stairs 100 in order to maintain the center of gravity of the robot 2 in a fixed position relative to the pair of legs 32, 33. Therefore, whether the robot 2 is holding a package 150 or another item (e.g., for delivery), or the robot 2 is moving without holding anything, the disclosed height change detection system 4 is configured so that the center of gravity does not change as the robot 2 traverses the structure. This can be understood as mimicking human behavior, where arm and leg movements can correspond to a fixed center of gravity for stability purposes.

[0040] Additionally, as described above, typical systems often require the use of separate sensors in order to allow the robot to ascend and descend relative to the structure. However, in one example embodiment of the disclosed concept, adjustments to the operating system 30 may be performed based on height changes (e.g., relative to the structure 100, 200, 300, 400) without a separate sensor spaced apart from the robot (e.g., a sensor coupled to the ceiling of a house). That is, the fused data provided by the camera 20 and the lidar sensor 22 may together allow for proper mobility of the robot 2 without the use of separate sensors. However, in one example, separate sensors may be used to further enhance the ability of the robot 2 to ascend and descend in a relatively efficient manner without tipping over.

[0041] Figure 7 Corresponding to an example process 500 in which altitude is classified by the altitude change detection system 4 of the disclosed concept. The example process 500 may include collecting camera data 502 and lidar sensor data 504 and fusing the data at step 506. In addition, the process may include detecting vertical edges 508 and detecting horizontal edges 510, and then classifying each as an ascending altitude 512 or a descending altitude 514.

[0042] Figure 8Corresponding to an example process 600 in which edges 102, 104, 106, 108, 202, 204, 206, 302, 304, 402 are detected by the height change detection system 4 of the disclosed concept. As shown, the raw point cloud data 602 and the IMU data 604 can be combined to produce rotated and filtered point cloud data 606. Subsequently, the horizontal edge 608 and the vertical edge 610 of the structure can be determined based on the corresponding horizontal point 612 and vertical point 614.

[0043] Fig. 9 7. An example algorithm 700 corresponding to one that may be employed by the processor 12 of the height change detection system 4 of the disclosed concept. The algorithm 700 may begin by employing edge points 702 to generate an edge image 704. The edge image 704 may be used to generate XY projected edge segments 706, which in turn may be used to generate 3D edges 708. Parallel edges of the 3D edges 708 may then be grouped together so that discrete heights may be counted at step 710. After step 710, stairs, cliffs, and boundaries 712, 714, 716 may be identified. Based on the identification of stairs 712, stair lip 718 may be identified, and the location, step height, depth, and count of the stairs may then be modeled at step 720. Additionally, the location and height of cliff 714 may be determined at step 722, and the location of boundary 716 may be determined at step 724.

[0044] Fig.10 800 corresponds to the process by which data from the camera 20 and the lidar sensor 22 are fused. As shown, the ROS subscriber node 806 receives the camera data 802 and the lidar sensor data 804 corresponding to the uplink altitude 810 and the downlink altitude 812, and then presents them into tags for later use by the ROS publisher node 808. The tags are sent to the rest of the operating system as ROS messages 814 that can provide altitude classification, altitude position, and altitude measurement values.

[0045] Fig.11 The chain 900 that an altitude tag goes through before being used to notify a low-level control is shown. The chain 900 includes an altitude message 902, a ROS subscriber node 904, a robot control module 906, a ROS publisher node 908, and a control step 910.

[0046] Fig.12 An isometric view of a quadruped robot 1002 and its height change detection system 1004 is shown in accordance with another non-limiting embodiment of the disclosed concepts. Fig.13A and Fig. 13BA wheeled robot 1102 and its height change detection system 1104 according to another non-limiting embodiment of the disclosed concepts are shown. It should be understood that the height change detection system 1004, 1104 of the robot 1002, 1102 can function substantially the same as the height change detection system 4 (e.g., a sensor suite can be employed to scan the edge of a structure to allow for classification of height changes, and in response, the operating system is adjusted so that the robot 1002, 1102 can traverse the structure or move away from the structure such as a cliff). In addition, the algorithms employed by the processors of the robots 2, 1002, 1102 can be platform-independent, so that they can be employed on any of the robots 2, 1002, 1102, as long as they are equipped with at least one sensor that provides point cloud data.

[0047] about Fig.13A and Fig. 13B As shown in the figure, the robot 1102 includes wheels, a top hat 1105, and a body 1107 coupled to the top hat 1105 and the wheels. The top hat 1105 can be configured to hold the package while the robot 1102 traverses the structure. It should be understood that the robot 1102 includes at least one actuated joint including a wheel, and when the actuated joint is at an upward angle ( Fig.13A ) or downward angle ( Fig. 13B ) while traversing a structure, the robot 1102 can maintain stability during ascent or descent by determining the position of the top cap 1105 relative to the body 1107 (as determined by a processor of the robot 1102). In addition, data processed by the processor of the height change detection system 1104 can be used to create a calibration for the robot 1102, which can be used during navigation on the structure to maintain stability during ascent ( Fig.13A ) or climb down ( Fig. 13B ). Separate calibrations can be used for ascending and descending, and for indoor and outdoor climbing (e.g., because outdoor stairs are generally less steep than indoor stairs).

[0048] It should also be understood that data processed by the processor 12 of the robot 2 (e.g., for a legged robot, where the actuated joints include the legs 32, 33) can be used to create a calibration for the robot 2 that can be used during navigation on a structure while maintaining stability when climbing up or down. For example, when the robot 2 is carrying a package of a certain weight, in order for it to maintain stability while performing a climbing task, the robot 2 must perform some slight displacement of the package load forward or backward (in addition to maintaining joint torque on the legs 32, 33 so as not to lose stability during climbing).

[0049] Thus, it should be appreciated that the method according to the disclosed concept includes: providing the robot 2, 1002, 1102 with an operating system 30 and a sensor 20, 22 electrically connected to the operating system 30; detecting the edge 102, 104, 106, 108, 202, 204, 206, 302, 304, 402 of the structure 100, 200, 300, 400 with the sensor 20, 22; classifying the height change associated with the edge 102, 104, 106, 108, 202, 204, 206, 302, 304, 402; and adjusting the operating system 30 based on the height change. The method may further include: providing the robot 2 with a pair of legs 32, 33; determining an ascending angle or a descending angle corresponding to the height change; causing the pair of legs 32, 33 to traverse the structure 100, 200, 300, 400 at the ascending angle or the descending angle, wherein the center of gravity remains fixed relative to the pair of legs 32, 33. In addition, the method may also include detecting edges 102, 104, 106, 108, 202, 204, 206, 302, 304, 402 of the structure 100, 200, 300, 400 using the camera 20 and the lidar sensor 22, and fusing data from the lidar sensor 22 with data from the camera 20 to classify height changes associated with the edges 102, 104, 106, 108, 202, 204, 206, 302, 304, 402.

[0050] The processor 12 of the robot 2 (e.g., and the processors of the robots 1002, 1102) may be a commercially available general purpose processor, such as a processor from or The memory 14 of the robot 2 (e.g., and the memory of the robots 1002, 1102) may be a non-transitory computer-readable memory storing program code, and may include any one or a combination of volatile memory elements (e.g., dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), etc.), and may include any one or more non-volatile memory elements (e.g., erasable programmable read-only memory (EPROM), flash memory, electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), etc.).

[0051] In the above disclosure, reference has been made to the accompanying drawings forming a part thereof, which illustrate specific embodiments in which the present disclosure may be practiced. It should be understood that other embodiments may be utilized and structural changes may be made without departing from the scope of the present disclosure. References in this specification to "one embodiment," "embodiment," "example embodiment," etc. indicate that the described embodiment may include specific features, structures, or characteristics, but each embodiment may not necessarily include the specific features, structures, or characteristics. In addition, such phrases do not necessarily refer to the same embodiment. In addition, when features, structures, or characteristics are described in conjunction with an embodiment, whether or not explicitly described, those skilled in the art will recognize such features, structures, or characteristics in conjunction with other embodiments.

[0052] In addition, where appropriate, the functions described herein may be performed in one or more of the following: hardware, software, firmware, digital components, or analog components. Certain terms are used throughout the specification and claims to refer to specific system components. As will be appreciated by those skilled in the art, components may be referred to by different names. This document is not intended to distinguish between components that have different names but the same function.

[0053] It should also be understood that the word "example" as used herein is intended to be non-exclusive and non-limiting in nature. More specifically, the word "example" used herein indicates one of several examples, and it should be understood that no undue emphasis or preference is placed on the specific example described.

[0054] Furthermore, the operations described herein should always be implemented and / or performed in accordance with the user manual and safety guidelines.

[0055] Additionally, with respect to the processes, systems, methods, heuristics, etc. described herein, it should be understood that although the steps of such processes, etc. have been described as occurring according to a certain ordered sequence, such processes may be practiced with the described steps performed in an order different than that described herein. It should also be understood that certain steps may be performed simultaneously, other steps may be added, or certain steps described herein may be omitted. In other words, the descriptions of processes herein are provided for the purpose of illustrating various embodiments and should in no way be construed as limiting the claims.

[0056] Therefore, it should be understood that the above description is intended to be illustrative rather than restrictive. Upon reading the above description, many embodiments and applications other than the examples provided will be apparent. The scope should not be determined with reference to the above description, but should be determined with reference to the entire scope of the appended claims and equivalents to which such claims are entitled. It is anticipated and expected that the technology discussed herein will develop in the future, and the disclosed systems and methods will be incorporated into such future embodiments. In short, it should be understood that the present application is capable of modification and change.

[0057] Unless explicitly indicated to the contrary in this article, all terms used in the claims are intended to be given their ordinary meanings as understood by the skilled person described herein. Specifically, unless the claims state an explicit limitation to the contrary, the use of singular articles such as "one", "the", "said" and the like should be interpreted as describing one or more of the indicated elements. Unless otherwise specifically stated or understood in other ways within the context when used, conditional language such as, in particular, "can", "may", "can" or "may" is generally intended to express that certain embodiments may include certain features, elements and / or steps, while other embodiments may not include certain features, elements and / or steps. Therefore, such conditional language is generally not intended to imply that one or more embodiments require each feature, element and / or step in any way.

[0058] According to an embodiment, the memory further comprises instructions which, when executed by the processor, cause the processor to perform operations comprising: adjusting a tilt angle of the sensor in response to detecting the edge.

[0059] According to the present invention, a method includes: providing a robot with an operating system and a sensor electrically connected to the operating system; detecting an edge of a structure with the sensor; classifying a height change associated with the edge; and adjusting the operating system based on the height change.

[0060] In one aspect of the invention, adjusting the operating system is performed in real time upon detecting the edge of the structure.

[0061] In one aspect of the present invention, the robot has a center of gravity, wherein the method further comprises: providing the robot with a pair of legs, wherein the pair of legs are configured to be moved by the operating system; determining an upward angle or a downward angle corresponding to a change in height; causing the pair of legs to traverse the structure at the upward angle or the downward angle, wherein the center of gravity remains fixed relative to the pair of legs, and wherein adjustments to the operating system are performed before causing the pair of legs to traverse the structure.

[0062] In one aspect of the invention, the sensor is a lidar sensor, and wherein the method further comprises: providing the robot with a camera; detecting the edge of the structure using the camera; and fusing data from the lidar sensor with data from the camera to classify the height change associated with the edge.

Claims

1. A height change detection system for a robot having an operating system, the height change detection system comprising: sensor; a processor electrically connected to the sensor; and a memory comprising instructions that, when executed by the processor, cause the processor to perform operations comprising: detecting the edge of the structure using the sensor, classifying a height change associated with the edge, and The operating system is adjusted based on the altitude change. 2 . The height change detection system of claim 1 , wherein the adjustment of the operating system is performed in real time when the edge of the structure is detected. 3 . The height change detection system of claim 2 , wherein the height change is one of a number of ascending stairs, a number of descending stairs, a curb, and a cliff.

4. The height change detection system according to claim 3, wherein the sensor is one of a lidar sensor and a camera.

5. The height change detection system of claim 3, wherein the sensor is a lidar sensor, wherein the height change detection system further comprises a camera, and wherein the memory further comprises instructions which, when executed by the processor, cause the processor to perform operations comprising: detecting the edge of the structure using the camera; and fusing data from the lidar sensor with data from the camera to classify the height change associated with the edge. 6 . The height change detection system of claim 3 , wherein the memory further comprises instructions that, when executed by the processor, cause the processor to perform operations comprising adjusting a tilt angle of the sensor in response to detecting the edge.

7. A height change detection system according to claim 3, wherein the edge of the structure is a first edge, and wherein the memory also includes instructions, which when executed by the processor cause the processor to perform operations including: using the sensor to detect a second edge, a third edge and a fourth edge of the structure; classifying height changes associated with the second edge, the third edge and the fourth edge; and continuously adjusting the operating system in real time based on the height changes associated with the second edge, the third edge and the fourth edge.

8. A robot comprising: operating system; a sensor electrically connected to the operating system; a processor electrically connected to the sensor; and a memory comprising instructions that, when executed by the processor, cause the processor to perform operations comprising: detecting the edge of the structure using the sensor, classifying a height change associated with the edge, and The operating system is adjusted based on the altitude change.

9. The robot of claim 8, wherein the robot further comprises an actuated joint, the actuated joint comprising a wheel or a pair of legs, wherein the actuated joint is configured to be moved by the operating system, and wherein the memory further comprises instructions which, when executed by the processor, cause the processor to perform operations comprising: determining an ascending angle or a descending angle corresponding to the height change; and causing the actuated joint to traverse the structure at the ascending angle or the descending angle.

10. The robot of claim 9, wherein the actuation joint comprises the wheel, wherein the robot further comprises a body coupled to the wheel, and a top cap coupled to the body, and wherein causing the actuation joint to traverse the structure at the ascending angle or the descending angle comprises stabilizing the robot by determining a position of the top cap relative to the body during ascending or descending.

11. The robot of claim 9, wherein the robot has a center of mass, wherein the actuation joint comprises the pair of legs, and wherein traversing the structure by the actuation joint is performed while the center of mass of the robot remains fixed relative to the pair of legs.

12. The robot of claim 11, wherein adjusting the operating system based on the height change is performed without a separate sensor spaced apart from the robot.

13. The robot of claim 11, wherein adjusting the operating system is performed prior to causing the pair of legs to traverse the structure.

14. The robot of claim 11 further comprising a body coupled to the pair of legs, and wherein the adjustments to the operating system are performed in real time upon detection of the edge of the structure.

15. The robot according to claim 14, wherein the robot is one of a biped robot and a quadruped robot.

Images