Pose determination method, apparatus, and autonomous mobile device

Abstract

The application provides a pose determination method and device and an autonomous mobile device, and relates to the technical field of artificial intelligence. The method comprises the following steps: when the autonomous mobile device is located in an indoor parking lot, determining whether the autonomous mobile device is located in an inter-floor transition area according to a road slope where the autonomous mobile device is currently located; when it is determined that the autonomous mobile device is located in the inter-floor transition area, acquiring real-time motion data through an internal detection module of the autonomous mobile device; and determining a first pose of the autonomous mobile device according to the real-time motion data. In this way, when the autonomous mobile device is located in a multi-floor parking lot for mapping, the problem of excessively large pose estimation error caused by scene degradation in the inter-floor transition area can be effectively avoided.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to a pose determination method, apparatus, and autonomous mobile device. Background Technology

[0002] With the rapid development of autonomous driving technology, autonomous parking functions have gradually become more widespread. Autonomous parking helps reduce the skill requirements for drivers, especially in congested parking lots, effectively mitigating the risk of accidents during parking. However, signal blockage is common in parking lots, making it difficult for autonomous mobile devices (such as vehicles with autonomous driving modes) to achieve location tracking using absolute positioning methods like GPS (Global Positioning System).

[0003] To improve the positioning accuracy of autonomous mobile devices in environments with signal obstruction, related technologies have provided solutions using SLAM (Simultaneous Localization and Mapping) technology. In the scenario of a multi-story parking garage, autonomous mobile devices also need to perform positioning and mapping in the inter-floor passageways between different floors. However, existing SLAM technologies are prone to scene degradation in the inter-floor passageway areas, resulting in inaccurate estimation of the autonomous mobile device's pose. Summary of the Invention

[0004] In view of this, embodiments of this application provide a pose determination method, apparatus, autonomous mobile device, electronic device, and computer-readable storage medium, aiming to provide a technical solution that can accurately estimate the pose of an autonomous mobile device for the special scenario of the inter-level transition area in a multi-level parking lot.

[0005] The first aspect of this application provides a pose determination method, comprising: when an autonomous mobile device is located in an indoor parking lot, determining whether the autonomous mobile device is located in an inter-level transition area based on the current road slope of the autonomous mobile device; when it is determined that the autonomous mobile device is located in an inter-level transition area, acquiring real-time motion data through the internal detection module of the autonomous mobile device; and determining a first pose of the autonomous mobile device based on the real-time motion data, wherein the first pose is used to construct map information of the inter-level transition area.

[0006] In some embodiments, the method further includes: measuring the pitch angle of the autonomous mobile device through an internal detection module; when it is determined that the pitch angle exceeds a preset angle range, acquiring first environmental information collected by the external detection module of the autonomous mobile device; and calculating the road slope based on the first environmental information.

[0007] In some embodiments, the inter-floor transition area is located between the first and second floors of the indoor parking lot. The pose determination method further includes: determining that the autonomous mobile device enters the inter-floor transition area through a first exit among a plurality of exits on the first floor; determining, according to a preset topology map, a first entrance among a plurality of entrances on the second floor where the autonomous mobile device leaves the inter-floor transition area, wherein the topology map includes the correspondence between the plurality of exits on the first floor and the plurality of entrances on the second floor; and determining a second pose of the autonomous mobile device based on the position of the first entrance on the second floor, wherein the second pose is used for positioning of the autonomous mobile device.

[0008] In some embodiments, determining the second pose of the autonomous mobile device based on the location of the first entrance on the second floor includes: determining local map information near the first entrance in a pre-built map of the second floor based on the location of the first entrance on the second floor; acquiring second environmental information collected by the external detection module of the autonomous mobile device when it is determined that the autonomous mobile device has arrived at the first entrance on the second floor; and determining the second pose of the autonomous mobile device based on the local map information and the second environmental information.

[0009] In some embodiments, the external detection module includes a ranging sensor, and the second environmental information includes point cloud information. Specifically, determining local map information near the first entrance in pre-constructed map information of the second floor based on the location of the first entrance in the second floor includes: determining a local point cloud contour near the first entrance in pre-constructed global point cloud contour of the second floor based on the location of the first entrance in the second floor; and determining a second pose of the autonomous mobile device based on the local map information and the second environmental information includes: matching the local point cloud contour with the second environmental information to determine the second pose of the autonomous mobile device.

[0010] In some embodiments, the indoor parking lot includes multiple floors, and the pose determination method further includes: determining that the autonomous mobile device enters the inter-floor transition area from the first floor among the multiple floors; determining the movement direction of the autonomous mobile device in the inter-floor transition area according to the road slope, wherein the movement direction includes an uphill direction or a downhill direction; determining the target floor where the autonomous mobile device is located when leaving the inter-floor transition area according to the movement direction, wherein the target floor is a floor higher than the first floor or a floor lower than the first floor; and determining a second pose of the autonomous mobile device according to pre-constructed map information of the target floor, wherein the second pose is used for positioning of the autonomous mobile device.

[0011] A second aspect of this application provides a pose determination device, comprising: a judgment module, configured to determine whether an autonomous mobile device is located in an inter-level transition area based on the current road slope of the autonomous mobile device when the autonomous mobile device is located in an indoor parking lot; an acquisition module, configured to acquire real-time motion data of the autonomous mobile device through an internal detection module of the autonomous mobile device when it is determined that the autonomous mobile device is located in an inter-level transition area; and a calculation module, configured to determine a first pose of the autonomous mobile device based on the real-time motion data, wherein the first pose is used to construct map information of the inter-level transition area.

[0012] A third aspect of this application provides an autonomous mobile device, comprising: an internal detection module disposed on the autonomous mobile device for acquiring real-time motion data of the autonomous mobile device; and a processor for implementing the pose determination method provided in the first aspect of this application.

[0013] A fourth aspect of this application provides an electronic device, comprising: a memory for storing computer instructions; and a processor for executing the computer instructions to implement the pose determination method provided in the first aspect of this application.

[0014] The fifth aspect of this application provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the pose determination method provided in the first aspect of this application.

[0015] The pose determination method, apparatus, autonomous mobile device, electronic device, and computer-readable storage medium provided in this application utilize the significant differences in road surface slope between different areas within a multi-level parking garage. Based on the road surface slope, it determines whether the autonomous mobile device is located in the inter-level transition zone. When the autonomous mobile device is determined to be in the inter-level transition zone, its pose is calculated using real-time motion data collected by an internal detection module. This method effectively avoids the problem of excessive pose estimation errors caused by scene degradation in the inter-level transition zone when mapping is performed on an autonomous mobile device within a multi-level parking garage.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not constitute a limitation on this application. Attached Figure Description

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. It should be understood that the accompanying drawings constitute a part of the specification and are used in conjunction with the embodiments of this application to explain this application, but do not constitute a limitation of this application. Unless otherwise stated, in the accompanying drawings, the same symbols and numbers generally represent the same steps or components.

[0018] Figure 1 The diagram shown is a schematic diagram of an exemplary pose determination system provided in an embodiment of this application.

[0019] Figure 2 The diagram shown is a schematic flowchart of a pose determination method provided in an embodiment of this application.

[0020] Figure 3 The diagram shown is a schematic flowchart of an exemplary pose determination method provided in an embodiment of this application.

[0021] Figure 4 The diagram shown is a flowchart of a pose determination method provided in another embodiment of this application.

[0022] Figure 5 The diagram shown is a schematic representation of a topology map in one embodiment of this application.

[0023] Figure 6 The diagram shown is a schematic flowchart of an exemplary pose determination method provided in an embodiment of this application.

[0024] Figure 7 The diagram shown is a flowchart of a pose determination method provided in another embodiment of this application.

[0025] Figure 8 The diagram shown is a schematic diagram of a pose determination device provided in an embodiment of this application.

[0026] Figure 9 The diagram shown is a schematic diagram of an autonomous mobile device provided in an embodiment of this application.

[0027] Figure 10 The diagram shown is an exemplary electronic device provided in an embodiment of this application. Detailed Implementation

[0028] To enable autonomous parking in parking lots, autonomous mobile devices need to be able to navigate autonomously in narrow and congested spaces. High-precision positioning is a crucial prerequisite for autonomous navigation. However, indoor parking lots (especially underground ones) often suffer from severe signal obstruction, rendering absolute positioning technologies like GPS unusable and causing conventional navigation and positioning solutions to fall short of the requirements for autonomous navigation.

[0029] Against this backdrop, SLAM technology has become the preferred tool for solving the positioning accuracy problem in parking lots. SLAM technology relies on environmental information collected by external sensors (cameras, LiDAR, etc.) on autonomous mobile devices to continuously perceive changes in the surrounding environment and infer changes in the position and attitude of the autonomous mobile device itself, thereby achieving high-precision positioning and mapping. When an autonomous mobile device moves within the same floor of a parking lot, only pose changes in three degrees of freedom (x, y, yaw) in a plane are involved. Therefore, in this case, a mature 2D SLAM solution for indoor structured environments can usually be used for map construction.

[0030] However, in multi-story parking garages with multiple floors, autonomous mobile devices need to traverse inter-floor transition zones (sloping passageways between floors) when moving between different floors. In these transition zones, due to the significant ground slope, the pose of the autonomous mobile device can change in all six degrees of freedom (x, y, z, roll, pitch, yaw). Therefore, the 2D planar assumption no longer holds in these transition zones, and 2D SLAM schemes cannot meet the positioning requirements.

[0031] To address this issue, one solution in the interlayer transition region is to employ 3D SLAM. This involves using a ranging sensor on the autonomous mobile device to scan the environment, collect spatial features, and obtain 3D point cloud information of the surrounding space. The pose of the autonomous mobile device at that moment can then be determined based on this 3D point cloud information. However, the inventors found that because the environment in the interlayer transition region often lacks distinct features and the spatial shape remains consistently similar, scene degradation easily occurs during 3D SLAM, making it difficult to accurately estimate the pose of the autonomous mobile device. Furthermore, implementing this solution requires collecting a large amount of point cloud data, which increases the computational cost of the autonomous mobile device and fails to meet the requirements for lightweight design.

[0032] In view of this, embodiments of this application provide a pose determination method. When it is determined that an autonomous mobile device is located in the inter-level transition area of ​​a multi-level parking lot, the pose of the autonomous mobile device is calculated based on the real-time motion data of the autonomous mobile device obtained by the internal detection module. This effectively avoids the problem of excessive pose estimation error caused by scene degradation in the inter-level transition area. Furthermore, this pose determination method eliminates the need for feature extraction and matching, effectively saving computational costs.

[0033] Exemplary System

[0034] Figure 1The diagram shown is an exemplary pose determination system 100 provided in an embodiment of this application. The pose determination system 100 may include an autonomous mobile device 110, a processing device 120, and a storage device 130.

[0035] The autonomous mobile device 110 can be a vehicle or robot equipped with an autonomous driving mode. In the embodiments of this application, the autonomous mobile device 110 may be equipped with an internal detection module 111 for collecting its own real-time motion data and an external detection module 112 for collecting 3D point cloud information of the environmental space. The internal detection module 111 may include, for example, a wheel speed encoder, an inertial measurement unit (IMU), or other internal sensing sensors; the external detection module 112 may include, for example, a mechanically rotating 3D LiDAR, a solid-state / semi-solid-state 3D LiDAR, a binocular vision sensor, or a structured light sensor (e.g., an RGB-D camera), or other ranging sensors.

[0036] The processing device 120 can communicate with the internal detection module 111 and the external detection module 112 to receive real-time motion data of the autonomous mobile device and / or 3D point cloud information of the surrounding environment of the autonomous mobile device, and perform the required processing actions based on this information. For example, the processing device 120 can be used to calculate the pose of the autonomous mobile device 110, to construct a map of the environment in which the autonomous mobile device 110 is located, and to perform information fusion of various sensors, etc.

[0037] Storage device 130 can be communicatively connected to processing device 120, so that processing device 120 can retrieve corresponding data from storage device 130 according to the processing action to be performed. Storage device 130 can also be used to receive and store data processed by processing device 120.

[0038] It should be understood that the processing device 120 can be a processor or a server, or an electronic device with processing capabilities such as a computer, mobile phone, tablet computer, or in-vehicle computer. The processing device 120 can be located in the cloud or locally on the autonomous mobile device 110, as long as it can communicate with the various functional modules on the autonomous mobile device 110. The storage device 130 can be a server device located in the cloud or a local storage medium, etc. The embodiments of this application do not limit the specific implementation of the processing device 120 and the storage device 130.

[0039] Exemplary methods

[0040] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments in this application are within the scope of protection of this application.

[0041] Figure 2 The diagram shown is a schematic flowchart of a pose determination method provided in an embodiment of this application. This method can, for example, be derived from... Figure 1 The processing device 120 in the pose determination system 100 shown executes the operation. For example... Figure 2 As shown, the method may include the following:

[0042] S210: When the autonomous mobile device is located in an indoor parking lot, determine whether the autonomous mobile device is located in the inter-level transition area based on the current road slope of the autonomous mobile device.

[0043] In multi-level parking garages (e.g., above-ground or underground multi-level parking garages), the driving environment of autonomous mobile devices includes the area within each floor and the transition area between floors. Typically, the surface of the area within a floor is relatively flat, while the slope of the transition area between floors is significantly greater than that within the floor. Therefore, when an autonomous mobile device is driving in a multi-level parking garage, it can determine whether it is currently in a transition area between floors based on the slope of the surface it is on.

[0044] To determine the current road slope of an autonomous mobile device, one approach is to directly acquire information about the surrounding environment using an external detection module mounted on the device, and then determine the road slope based on this information. However, this method consumes significant computational resources and is inefficient. Therefore, some implementations utilize an internal detection module within the autonomous mobile device to measure its own inertial information, such as the pitch angle, to determine the current road slope.

[0045] However, in certain special circumstances, such as when an autonomous mobile device traverses obstacles or rough terrain, its attitude may change significantly. In such cases, relying solely on the pitch angle of the autonomous mobile device to determine whether the current area is a transition zone between layers may lead to incorrect judgments.

[0046] To solve this problem, in such Figure 3 In one preferred implementation shown, the road slope where the autonomous mobile device is currently located can be determined by the following steps before performing the above-described S210.

[0047] S311: Measure the pitch angle of the autonomous mobile device through an internal detection module.

[0048] The internal detection module located in an autonomous mobile device may include an inertial measurement unit (IMU), which can be used to measure the inertial information of the autonomous mobile device itself, such as pitch angle, roll angle and yaw angle.

[0049] S312: When it is determined that the pitch angle exceeds the preset angle range, the first environmental information collected by the external detection module of the autonomous mobile device is obtained.

[0050] An external detection module mounted on the autonomous mobile device is used to acquire information about the external environment near the autonomous mobile device. The external detection module may include, for example, a ranging sensor to acquire point cloud information of the external environment space. The ranging sensor may include a lidar (e.g., a mechanically rotating 3D lidar, a solid-state / semi-solid-state 3D lidar) or a structured light camera, etc.

[0051] It's understandable that when moving within a layer of terrain with almost no slope, the pitch angle of the autonomous mobile device is relatively small. Therefore, if the measured pitch angle of the autonomous mobile device is less than a preset range, it can be directly determined that the autonomous mobile device is located within the layer, eliminating the need for slope detection and thus saving computational resources. See the following formula for details:

[0052]

[0053] In the formula, This represents the state of the autonomous mobile device; hold, down, and up represent the possible states of the autonomous mobile device, namely, within the layer, downhill, and uphill, respectively. The pitch angle is represented as the vehicle's pose. This is represented as the threshold for angle determination; This represents the operation of taking the absolute value of the angle.

[0054] As an example, the preset angle range can be set to -0.0873 radians to 0.0873 radians.

[0055] S313: The road slope is calculated based on the first environmental information.

[0056] Specifically, 3D point cloud information of the surrounding environment of the autonomous mobile device can be acquired using a ranging sensor as the current point cloud frame, and ground points can be extracted from the current point cloud frame. Further, by performing plane fitting on the ground points, the normal vector of the fitted plane is obtained. Then, by calculating the angle between the normal vector of the fitted plane and the normal vector of the horizontal plane, the road surface slope value can be obtained. The formula for calculating the road surface slope is shown below.

[0057]

[0058]

[0059] In the formula, n is the normal vector of the fitted plane, [0,0,1] is the normal vector of the horizontal plane, θ is the angle between the normal vector of the fitted plane and the normal vector of the horizontal plane, ||n|| represents taking the 2 norm of vector n, and slope is the road slope value.

[0060] In a preferred embodiment, the CSF filtering algorithm can be used to filter the point cloud data, which can improve efficiency while ensuring accuracy. Furthermore, the RANSAC method can be used to perform plane fitting on the obtained ground points.

[0061] According to the above implementation method, the pitch angle of the autonomous mobile device is first detected. When the pitch angle exceeds a preset range, the road slope is calculated based on the environmental point cloud information collected by the ranging sensor. This road slope is then used to determine the area where the autonomous mobile device is located in subsequent steps. This method effectively reduces the computational cost of road slope determination. By combining pitch angle determination and slope determination, the accuracy of determining the area where the autonomous mobile device is located is improved while also saving computational resources.

[0062] After determining the current road slope of the autonomous mobile device, it can be used to determine whether the area where the autonomous mobile device is located is an interlayer transition zone.

[0063] Specifically, if the absolute value of the detected road surface slope is less than the slope threshold, the autonomous mobile device can be determined to be located within the layer; if the absolute value of the detected road surface slope is greater than or equal to the slope threshold, the autonomous mobile device can be determined to be located in the inter-layer transition zone. As an example, the slope threshold can be set to 0.0875 radians. It should be understood that the slope threshold used to determine the current location can be set according to the actual application scenario, and the embodiments of this application do not impose detailed limitations on this.

[0064] S220: When it is determined that the autonomous mobile device is located in the interlayer transition area, real-time motion data is obtained through the internal detection module of the autonomous mobile device.

[0065] The internal detection module may include various intrinsic sensing sensors, installed on the autonomous mobile device, to measure the device's own state and obtain real-time motion data. Specifically, the real-time motion data may include the autonomous mobile device's speed, angular velocity, and acceleration.

[0066] As an example, the internal detection module may include a wheel speed encoder and an inertial measurement unit (IMU). The wheel speed encoder outputs the velocity of the autonomous mobile device along the x-axis and its angular velocity about the z-axis. In the autonomous mobile device's vehicle coordinate system, the x-axis points in the direction of travel, the y-axis points to the left, and the z-axis points upwards following the right-hand rule. The inertial measurement unit outputs the angular velocity and acceleration values ​​of the autonomous mobile device.

[0067] S230: Determines the first pose of the autonomous mobile device based on real-time motion data.

[0068] The first pose is used to construct map information for the transition area between layers.

[0069] Based on real-time motion data acquired by the internal detection module, the first pose of the autonomous mobile device in the interlayer transition region can be calculated. For example, dead reckoning (DR) positioning can be used to calculate the first pose of the autonomous mobile device. In the embodiments of this application, map information of the interlayer transition region can be constructed based on the first pose.

[0070] In a preferred embodiment, before executing S230, the initial pose of the autonomous mobile device can be obtained by fusing data from various intrinsically sensitive sensors included in the internal detection module. For example, the data can be processed through methods such as filtering optimization and nonlinear optimization.

[0071] For example, when it is determined that the autonomous mobile device is currently located in the interlayer transition region, an extended Kalman filter algorithm can be used to fuse the data from the wheel speed encoder and the inertial measurement unit to obtain the initial first pose of the autonomous mobile device. Based on this initial first pose, while the autonomous mobile device is located in the interlayer transition region, the first pose of the autonomous mobile device can be continuously calculated at a preset frequency based on the real-time motion data collected by the wheel speed encoder and the inertial measurement unit.

[0072] Optionally, when constructing a map of the interlayer transition area, the first pose of the autonomous mobile device obtained in S230 can be used to set constraints on the point cloud information collected by the ranging sensor, and the map can be constructed based on the constraints, thereby avoiding the scene degradation problem caused by relying solely on point cloud matching and improving the accuracy of the map of the interlayer transition area.

[0073] The pose determination method provided in this application utilizes the significant differences in road surface slope between different areas within a multi-level parking garage. It determines whether an autonomous mobile device is located in the inter-level transition zone based on the road surface slope, and when the autonomous mobile device is determined to be in the inter-level transition zone, its pose is calculated using real-time motion data collected by an internal detection module. This approach effectively avoids the problem of excessive pose estimation errors caused by scene degradation in the inter-level transition zone when mapping is performed on an autonomous mobile device within a multi-level parking garage.

[0074] Figure 4 The diagram shown is a flowchart illustrating a pose determination method according to another embodiment of this application. This method can, for example, be derived from... Figure 1 The processing device 120 in the pose determination system 100 shown executes the operation.

[0075] Typically, when using pre-built maps for localization in multi-story parking garages, the independence of maps between different floors necessitates re-localization based on the map of the next floor when an autonomous mobile device moves from one floor to another. For example, when an autonomous mobile device moves from the first floor to the second floor of a multi-story parking garage, it needs to be re-localized based on the map of the second floor to determine its pose within that floor. However, since the maps of the second floor and the first floor are independent, the pose cannot be calculated based on prior information upon entering the second floor; instead, the autonomous mobile device must be re-localized using the global map of the second floor. This approach involves a large amount of matching calculations, increasing the computational load of localization and impacting the real-time performance of the system.

[0076] In view of this, such as Figure 4 As shown, in Figure 2 or Figure 3 Based on the embodiments shown, another embodiment of this application provides a pose determination method that may further include the following:

[0077] S410: Determine that the autonomous mobile device enters the inter-floor transition area via the first exit among multiple exits on the first floor.

[0078] Multi-level parking garages typically consist of at least two floors, with adjacent floors generally connected by an inter-floor transition area. For example, an autonomous mobile device can travel from the first floor to the second floor via this transition area. Each floor may include exits and entrances connecting to adjacent floors; the inter-floor transition area connects adjacent floors through these exits and entrances. It is understood that when a floor includes multiple exits and / or entrances, each exit or entrance has a spatial correspondence with the entrances or exits of adjacent floors. Specifically, if an autonomous mobile device enters the inter-floor transition area from the first exit on the first floor (e.g., exit A on the basement level) and moves to the second floor (e.g., basement level two), it can enter the second floor from the first entrance on the second floor (e.g., entrance A on basement level two).

[0079] S420: Based on the preset topology map, determine the first entrance among multiple entrances on the second floor when the autonomous mobile device leaves the inter-floor transition area.

[0080] The topology map includes the correspondence between multiple exits on the first floor and multiple entrances on the second floor.

[0081] As mentioned earlier, considering the spatial correspondence between entrances and exits on adjacent floors, in this embodiment, a topology map can be pre-established based on this correspondence to indicate the connectivity between entrances and exits on each floor of the multi-story parking garage. According to the topology map, when it is determined that the autonomous mobile device enters the inter-floor transition area through the first exit on the first floor, it can be determined that the entrance where the autonomous mobile device enters the second floor is the second entrance on the second floor corresponding to the first exit on the first floor.

[0082] As an example, Figure 5 The image shown is a schematic diagram of a topological map. Figure 5 As shown, in some implementations, topological maps can consist of nodes and edges. This type of topological map primarily reflects the connectivity between nodes, without needing to express the precise location of the map. It is simple to construct and, as a more compact representation, avoids the interference caused by excessive detail in the map.

[0083] It should be understood that, since the entrance and exit of a floor may overlap in space in reality, in this embodiment, "entrance" and "exit" are terms defined only according to actual function and are not intended to express that "entrance" and "exit" are located in two different locations in space.

[0084] S430: Determine the second pose of the autonomous mobile device based on the location of the first entrance on the second floor.

[0085] The second pose is used for positioning of autonomous mobile devices.

[0086] Once it is determined that the autonomous mobile device enters the second floor through the first entrance, the location of the first entrance can be determined based on the map of the second floor, thus determining the location of the autonomous mobile device after entering the second floor. Using this location as prior information, the pose of the autonomous mobile device can be quickly determined based on the map of the second floor, achieving localization of the autonomous mobile device on the second floor.

[0087] In a preferred implementation, such as Figure 6 As shown, step S430 may specifically include the following:

[0088] S431: Based on the location of the first entrance on the second floor, determine the local map information near the first entrance from the pre-built map information of the second floor.

[0089] S432: When it is determined that the autonomous mobile device has arrived at the first entrance of the second floor, the second environmental information collected by the external detection module of the autonomous mobile device is obtained.

[0090] S433: Determine the second pose of the autonomous mobile device based on local map information and second environmental information.

[0091] As a specific example, the external detection module may include a ranging sensor, and the second environmental information may include point cloud information. In this case, S431 may include: determining a local point cloud profile near the first entrance in a pre-constructed global point cloud profile of the second floor based on the location of the first entrance in the second floor; S433 may include: matching the local point cloud profile with the second environmental information to determine the second pose of the autonomous mobile device.

[0092] The pose determination method provided in this application utilizes the spatial correspondence between entrances and exits on different floors of a multi-story parking lot. Based on this correspondence, a topological map is pre-constructed, thereby enabling the autonomous mobile device to quickly determine its approximate location on a new floor as prior information when it moves between different floors. This allows for rapid pose estimation of the autonomous mobile device on the new floor, significantly improving computational efficiency and reducing computational costs.

[0093] In another specific example, after the autonomous mobile device enters the inter-floor transition area, its direction of movement (uphill or downhill) can be determined first, and then the new floor the autonomous mobile device will reach can be determined based on its direction of movement. After determining that the autonomous mobile device has entered a new floor, the working map of the autonomous mobile device can be switched to the map of the new floor based on a preset topology map, and the approximate position of the autonomous mobile device on the map when entering the new floor can be determined.

[0094] For example, in Figure 5 In the topology map shown, L m,n This represents the nth entrance / exit on the m-th floor. According to this topology map, if the autonomous mobile device enters from the 1st exit on the 2nd floor (i.e., L...),... 2,1 After descending the slope through the inter-floor transition area to enter the first floor, it can be determined that the autonomous mobile device will be located at the first entrance (L) of the first floor. 1,1 (Near) This provides prior information for the positioning of autonomous mobile devices within a new floor, enabling rapid pose determination and improving positioning efficiency and accuracy.

[0095] Figure 7 The diagram shown is a schematic flowchart of a pose determination method provided in another embodiment of this application. Figure 7 As shown, Figure 2 The pose determination method provided in the illustrated embodiment may also include the following:

[0096] S710: Determines whether an autonomous mobile device enters an inter-floor transition zone from the first floor of a multi-floor system.

[0097] Specifically, the pitch angle of the autonomous mobile device and the road slope can be used to determine whether the autonomous mobile device has entered the inter-layer transition zone. The specific determination method can be referred to the description in the above embodiments, and will not be repeated here.

[0098] S720: Determine the direction of movement of autonomous mobile equipment in the interlayer transition area based on the road surface slope.

[0099] Specifically, the direction of movement can be uphill or downhill. When it is determined that the autonomous mobile device is located in the interlayer transition zone, the direction of movement of the autonomous mobile device can be determined based on the road surface slope, that is, whether the autonomous mobile device is in an uphill or downhill state.

[0100] As an example, if the detected road surface slope is greater than or equal to the slope threshold... This allows us to determine that the autonomous mobile device is moving uphill; if the detected road slope is less than or exceeds a slope threshold... If so, then the direction of movement of the autonomous mobile device is determined to be downhill. See the following formula for details.

[0101]

[0102] In the formula, V state This indicates the current state of the autonomous mobile device; hold indicates that the autonomous mobile device is not currently in the inter-layer transition area, down and up indicate that the autonomous mobile device is currently in a downhill and uphill state, respectively; s represents the current slope value of the road surface, and |s| represents taking the absolute value of the slope value; This indicates the slope threshold.

[0103] S730: Determine the target floor where the autonomous mobile device is located when it leaves the inter-floor transition area, based on the direction of movement.

[0104] The target floor is either a floor higher or lower than the first floor. After determining that the autonomous mobile device has entered the inter-floor transition zone, its direction of movement can be used to determine whether it is moving to a higher or lower floor. For example, the number of floors the autonomous mobile device has risen or fallen can be calculated based on the data collected by the internal detection module, thus allowing the determination of its current floor when it leaves the inter-floor transition zone.

[0105] S740: Determines the second pose of the autonomous mobile device based on the pre-built map information of the target floor.

[0106] The second pose is used for positioning of the autonomous mobile device. After the autonomous mobile device enters the target floor from the inter-floor transition area, its pose can be determined based on the pre-built map information of the target floor.

[0107] Preferably, the following can be used: Figure 6 The method described in the illustrated embodiment determines the entrance where the autonomous mobile device enters the target floor based on a preset topology map, thereby enabling the rapid determination of the autonomous mobile device's position and orientation within the target floor.

[0108] The pose determination method provided in this application, for the scenario of multi-story parking lots, can accurately determine the floor that the autonomous mobile device enters after moving between floors, and can quickly achieve repositioning in the new floor, providing a positioning scheme that significantly improves computational efficiency while ensuring high accuracy.

[0109] As a specific implementation method, in execution Figures 2 to 7 When using any of the pose determination methods provided in the embodiments of this application, the process of making a preliminary estimate of the pose of the autonomous mobile device and preprocessing the point cloud data may include the following:

[0110] First, the state equation and observation equation can be established in the following way.

[0111] a. Establish the vehicle coordinate system of the autonomous mobile device. The x-axis points in the forward direction of the autonomous mobile device, the y-axis points to the left of the autonomous mobile device, and the z-axis follows the right-hand rule, pointing upwards on the autonomous mobile device. It should be understood that this embodiment does not consider special cases such as sideslip or bumps of the autonomous mobile device, i.e., it ignores the velocity and acceleration of the autonomous mobile device along the y and z axes.

[0112] Based on this, the system state can be established as follows:

[0113]

[0114] b. Therefore, the state equations of the system can be established as follows:

[0115]

[0116] In the formula, X k X k+1 Let Q represent the system states at times k and k+1, respectively. k Let f(∙) represent the process noise at time k, and let f(∙) represent the nonlinear function.

[0117] If a constant acceleration / angular velocity model is used to predict the system state during the filter's prediction phase, then the nonlinear function f(∙) can be represented by the state transition matrix A. The diagonal elements of the state transition matrix A are 1s, and all off-diagonal elements are 0s, except for the elements described below.

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132] The inertial measurement unit (IMU) can output angular velocity and acceleration values, while the wheel speed encoder can output the velocity along the x-axis and the angular velocity around the z-axis. Based on this, the measurements from each sensor can be directly used as observations to correct the system's state.

[0133] c. Specifically, the system observation equations are established as follows:

[0134]

[0135] In the formula, Y k+1 For the system observations at time k+1, R k+1 Let h(∙) be the observation noise at time k+1, and h(∙) be the system observation function. It should be understood that in this embodiment, the measured value is used as the observation value, so the matrix representation of the observation function is an identity matrix.

[0136] Furthermore, considering that the output of the wheel speed encoder is a pulse signal, a simple mathematical model can be used to convert the pulse signal into the linear velocity of the left and right wheels of the autonomous mobile device. In this embodiment, a two-wheel differential speed model as shown in the following formula can be used to convert the linear velocity of the left and right wheels into the corresponding pose of the autonomous mobile device.

[0137]

[0138]

[0139] In the formula, v x ω represents the velocity of the autonomous mobile device along the x-axis. z v represents the angular velocity of the autonomous mobile device about the z-axis. r v represents the linear velocity of the right wheel of an autonomous mobile device. l ∆d represents the linear velocity of the left wheel of the autonomous mobile device, and ∆d represents the axle length between the left and right wheels of the autonomous mobile device.

[0140] After establishing the system's state equation and observation equation according to the above steps, the initial pose estimate of the autonomous mobile device can be obtained by using the update / correction step in the extended Kalman filter algorithm.

[0141] d. Based on the extended Kalman filter algorithm, data collected by multiple sensors on the autonomous mobile device are fused to make a preliminary estimate of the pose of the autonomous mobile device.

[0142] In this implementation, the autonomous mobile device can be an autonomous vehicle or a mobile robot, or other vehicle suitable for using SLAM technology. Various sensors can be included, for example, wheel speed encoders, inertial measurement units (IMUs), and ranging sensors. The ranging sensor can be any sensor capable of providing 3D point cloud information of the environmental space, such as a mechanically rotating 3D LiDAR, a solid-state / semi-solid-state 3D LiDAR, or a structured light camera.

[0143] e. Preprocess the acquired point cloud data.

[0144] Preprocessing can include operations such as distortion removal, time alignment, point cloud downsampling, and coordinate system transformation.

[0145] 1) Distortion removal

[0146] During the process of a ranging sensor acquiring a frame of environmental information, the autonomous mobile device inevitably moves, causing changes in its relative pose with the spatial environment (such as obstacles), which increases the measurement error of the ranging sensor. To reduce this measurement error caused by the movement of the autonomous mobile device, distortion removal processing can be performed on the point cloud before using the point cloud frame for calculation.

[0147] Specifically, based on the timestamp of each point in the point cloud frame and the pose of the autonomous mobile device at each time point provided by the pose fusion module, the pose of the autonomous mobile device at each time point corresponding to each point in the point cloud frame is calculated using a linear interpolation algorithm. See the following formula for details:

[0148]

[0149] In the formula, express The pose of the autonomous mobile device output by the time-position fusion module. This represents the proportion of time t to k within the time interval t to t+1, i.e. .

[0150] 2) Time alignment

[0151] Furthermore, point cloud frames from different times can be aligned to the same time frame, as shown in the following formula:

[0152]

[0153] In the formula, Indicates the current point cloud frame. One point, Indicates the first The pose of the autonomous mobile device obtained by point interpolation. This represents the inverse of the pose of the autonomous mobile device obtained by interpolating the first point in the current point cloud frame. This represents the coordinates of the point after distortion removal.

[0154] 3) Downsampling

[0155] To reduce computational load and save computational costs, voxel filtering can be applied to the point cloud frames output by the ranging sensor, thereby achieving downsampling.

[0156] Preferably, in the embodiments of this application, the voxel filtering method in the related art can be improved. Specifically, instead of calculating the centroid of all points falling in the current grid in the related art, a method of randomly selecting one point from all points as the representative value of the current grid is used for calculation.

[0157] Additionally, the minimum number of points in the downsampled point cloud frame can be set to control the voxel size.

[0158] 4) Coordinate system transformation

[0159] It should be understood that the research object in the embodiments of this application is an autonomous mobile device. To facilitate data processing, the point cloud in the sensor coordinate system can be transformed to the vehicle coordinate system of the autonomous mobile device. Specifically, the coordinate transformation matrix of the point cloud can be obtained through joint calibration of the ranging sensor and the inertial measurement unit (IMU).

[0160] It should be understood that the above steps can be combined with the steps or methods provided in the foregoing embodiments of this application. The specific combination can be designed by those skilled in the art according to actual needs, and the embodiments of this application do not limit this.

[0161] Exemplary device

[0162] Figure 8 The diagram shown is a schematic representation of a pose determination device 800 provided in an embodiment of this application. Figure 8 As shown, the pose determination device 800 provided in this application embodiment may include a judgment module 810, an acquisition module 820, and a calculation module 830.

[0163] The judgment module 810 can be used to determine whether the autonomous mobile device is located in the inter-level transition area based on the current road slope when the autonomous mobile device is located in an indoor parking lot; the acquisition module 820 can be used to acquire the real-time motion data of the autonomous mobile device through the internal detection module of the autonomous mobile device when it is determined that the autonomous mobile device is located in the inter-level transition area; the calculation module 830 can be used to determine the first pose of the autonomous mobile device based on the real-time motion data, wherein the first pose is used to construct the map information of the inter-level transition area.

[0164] In one embodiment, the pose determination device 800 may further include a measurement module for measuring the pitch angle of the autonomous mobile device via an internal detection module. The acquisition module 820 may also be used to acquire first environmental information collected by the external detection module of the autonomous mobile device when the pitch angle is determined to exceed a preset angle range. The calculation module 830 may also be used to calculate the road surface slope based on the first environmental information.

[0165] In one embodiment, the inter-floor transition area is located between the first and second floors of the indoor parking lot, and the pose determination device 800 further includes a determination module. The determination module can be used to perform the following steps: determining that the autonomous mobile device enters the inter-floor transition area via a first exit from among multiple exits on the first floor; determining, based on a preset topology map, a first entrance from among multiple entrances on the second floor where the autonomous mobile device leaves the inter-floor transition area, wherein the topology map includes the correspondence between multiple exits on the first floor and multiple entrances on the second floor; and determining a second pose of the autonomous mobile device based on the position of the first entrance on the second floor, wherein the second pose is used for positioning of the autonomous mobile device.

[0166] Furthermore, the determining module can also be used to determine local map information near the first entrance in the pre-built map information of the second floor based on the location of the first entrance in the second floor. Here, the acquiring module 820 can also be used to acquire the second environmental information collected by the external detection module of the autonomous mobile device when it is determined that the autonomous mobile device has arrived at the first entrance of the second floor; the calculating module 830 can also be used to determine the second pose of the autonomous mobile device based on the local map information and the second environmental information.

[0167] Furthermore, the external detection module may include a ranging sensor, and correspondingly, the second environmental information includes point cloud information. Specifically, the determination module may be used to determine a local point cloud contour near the first entrance in a pre-constructed global point cloud contour of the second floor based on the location of the first entrance in the second floor; the calculation module 830 may be used to match the local point cloud contour with the second environmental information to determine the second pose of the autonomous mobile device.

[0168] Optionally, as another embodiment, the indoor parking lot includes multiple floors. The determining module is further configured to perform the following steps: determining that the autonomous mobile device enters the inter-floor transition area from the first floor among the multiple floors; determining the direction of movement of the autonomous mobile device within the inter-floor transition area based on the road slope, wherein the direction of movement includes an uphill or downhill direction; and determining the target floor where the autonomous mobile device will be located when leaving the inter-floor transition area based on the direction of movement, wherein the target floor is a floor higher than or lower than the first floor. Simultaneously, the calculation module 830 can be configured to determine a second pose of the autonomous mobile device based on pre-constructed map information of the target floor. The second pose is used for positioning of the autonomous mobile device.

[0169] The pose determination device provided in this application utilizes the significant differences in road surface slope between different areas within a multi-level parking garage. It determines whether an autonomous mobile device is located in the inter-level transition zone based on the road surface slope. When the autonomous mobile device is determined to be in the inter-level transition zone, its pose is calculated using real-time motion data collected by an internal detection module. This method effectively avoids the problem of excessive pose estimation errors caused by scene degradation in the inter-level transition zone when mapping is performed on an autonomous mobile device within a multi-level parking garage.

[0170] It should be understood that the principles, functions, characteristics of the data used, processing methods for the data, and technical effects of all optional implementation methods of each module in the pose determination device 800 provided in this embodiment can be referred to the corresponding content in the exemplary method, and will not be repeated here.

[0171] Exemplary device

[0172] Figure 9 The diagram shown is of an autonomous mobile device 900 provided in an embodiment of this application. The autonomous mobile device 900 can be a vehicle with autonomous driving capabilities, or an electronic device such as a robot with autonomous mobility capabilities.

[0173] like Figure 9 As shown, the autonomous mobile device may include an internal detection module 910 and a processor 920. The internal detection module 910 is located on the autonomous mobile device and can acquire real-time motion data of the autonomous mobile device, such as data from sensors like wheel speed encoders and inertial measurement units (IMUs). Furthermore, the processor 920 can be used to execute the pose determination method provided in any of the above embodiments of this application to provide accurate pose estimation.

[0174] Figure 10 The diagram shown is an exemplary electronic device 1000 provided in an embodiment of this application. Figure 10As shown, the electronic device may include a processor 1010 and a memory 1020. The memory 1020 stores computer instructions, and the processor 1010 executes the computer instructions to implement the pose determination method provided in any of the above embodiments.

[0175] Exemplary computer-readable storage media

[0176] Other embodiments of this application also provide a computer-readable storage medium including computer instructions stored thereon, which, when executed by a processor, cause the processor to perform a pose determination method as provided in any of the above embodiments.

[0177] Alternatively, the computer storage medium can be any tangible medium, such as a floppy disk, CD-ROM, DVD, hard disk drive, or network media.

[0178] Exemplary computer program product

[0179] Other embodiments of this application also provide a computer program product, which includes instructions that, when executed by the processor of a computer device, enable the computer device to perform the pose determination method provided in any of the above embodiments of this application.

[0180] The block diagrams of the devices, equipment, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. Those skilled in the art will understand that these devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms, up to “including but not limited to,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0181] It should also be noted that in the apparatus, equipment, and methods of this application, each module or step can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions of this application.

[0182] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the foregoing aspects, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0183] The above description is for illustrative purposes only, illustrating and describing the technical solutions of this application. Furthermore, this description is not intended to limit the embodiments of this application to the scope of the forms disclosed above. Although several exemplary aspects and embodiments have been discussed above, those skilled in the art can readily derive other variations, modifications, alterations, additions, and sub-combinations based on the above content.

[0184] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A pose determination method, characterized in that, include: When the autonomous mobile device is located in an indoor parking lot, it is determined whether the autonomous mobile device is located in the inter-level transition area based on the current road slope of the autonomous mobile device. When it is determined that the autonomous mobile device is located in the interlayer transition area, real-time motion data is obtained through the internal detection module of the autonomous mobile device. The first pose of the autonomous mobile device is determined based on the real-time motion data, wherein the first pose is used to construct map information for the interlayer transition region. The inter-floor transition area is located between the first and second floors of the indoor parking garage. The pose determination method further includes: The autonomous mobile device is determined to enter the inter-floor transition area through the first exit among multiple exits on the first floor; Based on a preset topology map, the first entrance of the autonomous mobile device is located on the second floor among multiple entrances when it leaves the inter-floor transition area. The topology map includes the correspondence between multiple exits on the first floor and multiple entrances on the second floor. Based on the location of the first entrance on the second floor, a second pose of the autonomous mobile device is determined, wherein the second pose is used for positioning of the autonomous mobile device.

2. The pose determination method according to claim 1, characterized in that, Also includes: The pitch angle of the autonomous mobile device is measured by the internal detection module. When it is determined that the pitch angle exceeds the preset angle range, the first environmental information collected by the external detection module of the autonomous mobile device is obtained; The road surface slope is calculated based on the first environmental information.

3. The pose determination method according to claim 1, characterized in that, Determining the second pose of the autonomous mobile device based on the location of the first entrance on the second floor includes: Based on the location of the first entrance on the second floor, determine the local map information near the first entrance from the pre-constructed map information of the second floor; When it is determined that the autonomous mobile device has arrived at the first entrance of the second floor, the second environmental information collected by the external detection module of the autonomous mobile device is obtained; The second pose is determined based on the local map information and the second environmental information.

4. The pose determination method according to claim 3, characterized in that, The external detection module includes a ranging sensor, and the second environmental information includes point cloud information, wherein... Based on the location of the first entrance on the second floor, local map information near the first entrance is determined from the pre-built map information of the second floor, including: determining the local point cloud contour near the first entrance from the pre-built global point cloud contour of the second floor based on the location of the first entrance on the second floor. Determining the second pose based on the local map information and the second environment information includes: matching the local point cloud contour with the second environment information to determine the second pose.

5. The pose determination method according to claim 1, characterized in that, The indoor parking lot includes multiple floors, and the pose determination method further includes: Determine that the autonomous mobile device enters the inter-floor transition area from the first floor of the plurality of floors; Based on the road surface slope, the direction of movement of the autonomous mobile device in the interlayer transition area is determined, wherein the direction of movement includes an uphill direction or a downhill direction; Based on the direction of movement, the target floor where the autonomous mobile device is located when it leaves the inter-floor transition area is determined, wherein the target floor is a floor higher than the first floor or a floor lower than the first floor; Based on the pre-constructed map information of the target floor, a second pose of the autonomous mobile device is determined, wherein the second pose is used for positioning of the autonomous mobile device.

6. A pose determination device, characterized in that, include: The judgment module is used to determine whether the autonomous mobile device is located in the inter-floor transition area based on the current road slope of the autonomous mobile device when the autonomous mobile device is located in the indoor parking lot. The inter-floor transition area is located between the first floor and the second floor of the indoor parking lot. The acquisition module is used to acquire real-time motion data of the autonomous mobile device through the internal detection module of the autonomous mobile device when it is determined that the autonomous mobile device is located in the interlayer transition area; The calculation module is used to determine the first pose of the autonomous mobile device based on the real-time motion data, wherein the first pose is used to construct map information of the interlayer transition region; The determining module is configured to: determine whether the autonomous mobile device enters the inter-floor transition area via a first exit among multiple exits on the first floor; determine, based on a preset topology map, whether the autonomous mobile device is located at a first entrance among multiple entrances on the second floor when leaving the inter-floor transition area, wherein the topology map includes the correspondence between multiple exits on the first floor and multiple entrances on the second floor; and determine a second pose of the autonomous mobile device based on the position of the first entrance on the second floor, wherein the second pose is used for positioning of the autonomous mobile device.

7. An autonomous mobile device, characterized in that, include: An internal detection module, located on the autonomous mobile device, is used to acquire real-time motion data of the autonomous mobile device; A processor for executing the pose determination method according to any one of claims 1-5.

8. An electronic device, characterized in that, include: Memory, used to store computer instructions; A processor for executing the computer instructions to implement the pose determination method according to any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The device stores computer instructions, which, when executed by a processor, implement the pose determination method according to any one of claims 1-5.

Images