Navigation method and device for robot, robot and computer readable medium

Abstract

The embodiment of the invention discloses a navigation method and device for a robot, the robot, a computer readable medium and a program product. A specific embodiment of the method comprises the following steps: determining a target observation point according to current navigation information; for each candidate motion trail in the obtained candidate motion trail group, determining sight line information of a trail point on the candidate motion trail towards the target observation point; selecting a candidate motion track from the candidate motion track group according to the determined sight line information; and generating a target motion track according to the selected candidate motion track, and performing navigation according to the target motion track. The implementation mode is related to the intelligent robot, and active safety guarantee and smooth obstacle avoidance in the navigation process of the robot can be achieved.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of computer technology, and more particularly to robot navigation methods, apparatus, robots, computer-readable media, and program products. Background Technology

[0002] Currently, autonomous mobile robots used in scenarios such as homes and warehouses typically employ a single visual sensor (such as a camera) as the primary sensing device to achieve low-cost, lightweight navigation. However, due to the inherent limitation of the field of view, visual sensors are prone to creating large blind spots around the robot, which may lead to collisions or getting lost in complex dynamic environments.

[0003] To address the above issues, wide-angle or fisheye lenses are typically used to expand the field of view. Alternatively, multiple homogeneous or heterogeneous sensors (such as LiDAR, ultrasonic sensors, etc.) can be deployed around the robot, combined with fusion algorithms to achieve omnidirectional environmental perception. Another approach is to employ a combination of global and local path planning to achieve reactive navigation: avoiding obstacles and stopping if detours fail. Summary of the Invention

[0004] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0005] Some embodiments of this disclosure provide navigation methods, apparatus, robots, computer-readable media, and program products for robots to address the technical problems mentioned in the background section above.

[0006] In a first aspect, some embodiments of this disclosure provide a navigation method for a robot, including: determining a target observation point based on current navigation information; for each candidate motion trajectory in an acquired candidate motion trajectory group, determining line-of-sight information of trajectory points on the candidate motion trajectory toward the target observation point; selecting a candidate motion trajectory from the candidate motion trajectory group based on the determined line-of-sight information; generating a target motion trajectory based on the selected candidate motion trajectory; and navigating based on the target motion trajectory.

[0007] Optionally, determining the target observation point based on the current navigation information includes: determining a candidate point set based on the current navigation information, wherein the candidate point set includes at least one of the following candidate points: dynamic points, detection points, and preset guide points, wherein dynamic points include the location of dynamic obstacles, and detection points include blind zone boundary points; determining the navigation value information of each candidate point in the candidate point set; and selecting a candidate point from the candidate point set as the target observation point based on the determined navigation value information.

[0008] Optionally, the above method further includes: for each candidate motion trajectory in the candidate motion trajectory group, determining at least one of trajectory deviation information, collision risk information, and tracking feasibility information corresponding to the candidate motion trajectory, wherein the trajectory deviation information characterizes the difference between the candidate motion trajectory and the reference motion trajectory, the collision risk information characterizes the collision risk of the robot motion on the candidate motion trajectory, and the tracking feasibility information characterizes the feasibility of the robot motion tracking the target observation point on the candidate motion trajectory; and the above-mentioned selection of candidate motion trajectories from the candidate motion trajectory group based on the determined line-of-sight information includes: selecting candidate motion trajectories from the candidate motion trajectory group based on at least one of trajectory deviation information, collision risk information, tracking feasibility information, and line-of-sight information.

[0009] Optionally, the above method further includes: generating a tracking trajectory for tracking the target observation point based on the acquired robot motion parameters and target motion trajectory; and the above navigation based on the target motion trajectory includes: navigation based on the target motion trajectory and the tracking trajectory.

[0010] Optionally, the above method further includes: in response to detecting a trajectory update event during movement along the target trajectory, generating a new target trajectory based on the current navigation information, target observation point, candidate trajectory group, and line-of-sight information at the update time, and navigating based on the new target trajectory, wherein the update time includes the time when the trajectory update event was detected.

[0011] Secondly, some embodiments of this disclosure provide a navigation device for a robot, comprising: a determining unit configured to determine a target observation point based on current navigation information; the determining unit further configured to determine, for each candidate motion trajectory in an acquired group of candidate motion trajectories, line-of-sight information of trajectory points on the candidate motion trajectory toward the target observation point; a selecting unit configured to select a candidate motion trajectory from the group of candidate motion trajectories based on the determined line-of-sight information; and a navigation unit configured to generate a target motion trajectory based on the selected candidate motion trajectory, and to navigate based on the target motion trajectory.

[0012] Optionally, the aforementioned determining unit is further configured to: determine a candidate point set based on current navigation information, wherein the candidate point set includes at least one of the following candidate points: dynamic points, detection points, and preset guide points, wherein dynamic points include the location of dynamic obstacles, and detection points include blind zone boundary points; determine the navigation value information of each candidate point in the candidate point set; and select candidate points from the candidate point set as target observation points based on the determined navigation value information.

[0013] Optionally, the determining unit is further configured to: for each candidate motion trajectory in the candidate motion trajectory group, determine at least one of trajectory deviation information, collision risk information, and tracking feasibility information corresponding to the candidate motion trajectory, wherein the trajectory deviation information characterizes the difference between the candidate motion trajectory and the reference motion trajectory, the collision risk information characterizes the collision risk of the robot motion on the candidate motion trajectory, and the tracking feasibility information characterizes the feasibility of the robot motion tracking the target observation point on the candidate motion trajectory; and the selecting unit is further configured to: select a candidate motion trajectory from the candidate motion trajectory group based on at least one of the trajectory deviation information, collision risk information, and tracking feasibility information and the line of sight information.

[0014] Optionally, the navigation unit is further configured to: generate a tracking trajectory for tracking the target observation point based on the acquired robot motion parameters and target motion trajectory; and perform navigation based on the target motion trajectory and the tracking trajectory.

[0015] Optionally, the above apparatus further includes: an update unit configured to, in response to detecting a trajectory update event during movement along a target trajectory, generate a new target trajectory based on current navigation information, target observation point, candidate trajectory group, and line-of-sight information at the update time, and perform navigation based on the new target trajectory, wherein the update time includes the time when the trajectory update event is detected.

[0016] Thirdly, some embodiments of this disclosure provide a robot, including: a vision sensor, a processing module, and a control module; the processing module uses the vision sensor to determine a target observation point based on current navigation information; for each candidate motion trajectory in the acquired candidate motion trajectory group, it determines the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point; based on the determined line-of-sight information, it selects a candidate motion trajectory from the candidate motion trajectory group; it generates a target motion trajectory based on the selected candidate motion trajectory; and the control module navigates according to the target motion trajectory.

[0017] Optionally, the robot further includes a support device for supporting the vision sensor; and the processing module generates a tracking trajectory for tracking the target observation point based on the acquired robot motion parameters and target motion trajectory; and the control module performs navigation based on the target motion trajectory and the tracking trajectory.

[0018] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0019] Fifthly, some embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0020] The above embodiments of this disclosure have the following beneficial effects: the navigation methods for robots provided by some embodiments of this disclosure help to achieve forward-looking safety assurance and smooth guided navigation in low-cost, lightweight navigation scenarios.

[0021] Specifically, navigation methods that utilize wide-angle or fisheye lenses to expand the field of view typically require algorithms such as distortion correction, which increases the complexity of the navigation algorithm. Moreover, edge resolution and information reliability often deteriorate under ultra-wide field of view, thus affecting navigation accuracy. Navigation methods using multiple sensors significantly increase hardware costs and power consumption, and navigation using fusion algorithms requires addressing issues such as a sharp increase in complexity, making them difficult to apply in low-cost, lightweight navigation scenarios.

[0022] Furthermore, existing navigation methods that combine global and local path planning typically focus on immediate threats during movement, exhibiting a short-sighted behavioral pattern. Collisions heavily rely on sensors' timely detection of sudden obstacles in the near field, lacking predictive consideration of potential safety risks and global information gain. This results in abrupt and disjointed robot trajectories. Especially in situations such as observing behind a corner, navigation behavior exhibits a fragmented process of "movement-stagnation-perception-planning."

[0023] In response to the above situation, some embodiments of the present disclosure provide a navigation method for robots that determines observation points with high navigation value based on current navigation information and selects a motion trajectory with a good line of sight toward the observation point for robot navigation. That is, during the movement, key observation points are identified in a forward-looking manner and a motion path with a good field of vision is planned, so that the robot can continuously run on a path with sufficient field of vision information and controllable risks, thereby avoiding risks from the source.

[0024] Simultaneously, the visual sensor is used not only for environmental perception but also to continuously determine key observation points based on current navigation information to actively guide the robot to choose a movement trajectory with better field of vision. This means guiding the robot to move purposefully and safely through proactive observation. The existing perception-driven robot movement is transformed into perception-guided robot movement; that is, the existing "motion-perception" fragmented navigation mode is adjusted to a "perception-motion" collaborative navigation mode. This ensures that the robot can simultaneously detect its surrounding environment during smooth and continuous movement, thereby improving navigation intelligence and fluency. Moreover, this navigation method does not increase additional hardware costs or computational complexity, making it well-suited for low-cost, lightweight navigation scenarios. Attached Figure Description

[0025] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0026] Figure 1 This is a schematic diagram of an application scenario of a navigation method for a robot according to some embodiments of the present disclosure; Figure 2 This is a flowchart of some embodiments of the navigation method for robots according to the present disclosure; Figure 3 These are flowcharts of other embodiments of the navigation method for robots according to this disclosure; Figure 4 This is a flowchart of yet another embodiment of the navigation method for robots according to the present disclosure; Figure 5 These are schematic diagrams illustrating the structure of some embodiments of a navigation device for a robot according to the present disclosure; Figure 6 This is a schematic diagram of the structure of a robot suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0027] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0028] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0029] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0030] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0031] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0032] Before performing any of the operations involving the collection, storage, and use of user personal information (such as user profiles and user historical behavior) disclosed in this disclosure, the relevant organizations or individuals shall fulfill their obligations, including conducting personal information security impact assessments, informing personal information subjects, and obtaining prior authorization and consent from personal information subjects.

[0033] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0034] Figure 1 This is a schematic diagram of an application scenario of a navigation method for a robot according to some embodiments of the present disclosure.

[0035] exist Figure 1 In this application scenario, when robot 101 receives an instruction to retrieve lunch from the kitchen and bring it to the living room, it can determine the location of child 102 as the target observation point based on navigation information such as the current map of the house (e.g., the locations and relative positions of the balcony, bedroom, kitchen, living room, and bathroom), detected moving obstacles (i.e., child 102 playing in the house), and reference movement trajectory 103. Then, it determines the line-of-sight points on candidate movement trajectories 104 and 105 towards child 102, and selects candidate movement trajectory 104, which allows continuous observation of child 102, as the target movement trajectory, enabling robot 101 to retrieve lunch from the kitchen and bring it to the living room according to candidate movement trajectory 104.

[0036] It should be noted that, depending on the actual application scenario, the robot 101 described above can be of various types, such as a home robot, a warehouse robot, etc. Various devices or components can also be installed on the robot 101 according to actual needs, such as installing a gimbal to control the rotation of the vision sensor, etc. No specific limitations are made here.

[0037] It should be noted that the executor of the navigation method for robots provided in the embodiments of this disclosure is generally a robot (e.g., Figure 1 The robot 101 shown is used for navigation. The navigation method for the robot is generally executed by the robot 101, and correspondingly, the navigation device for the robot is generally installed in the robot 101.

[0038] It should be understood that Figure 1 The moving obstacles and candidate motion trajectories shown are merely illustrative. Depending on the implementation requirements, any number of moving obstacles and candidate motion trajectories can be included.

[0039] Continue to refer to Figure 2 The diagram illustrates a flow 200 of some embodiments of a navigation method for a robot according to the present disclosure. The navigation method for a robot includes the following steps: Step 201: Determine the target observation point based on the current navigation information.

[0040] In some embodiments, the execution subject of the above-described navigation method for robots (e.g. Figure 1 The robot 101 shown can first obtain current navigation information from its local device and / or other external devices connected to it. This current navigation information includes navigation information at the current time. The navigation information can include various information that can be used for robot navigation, and can be flexibly set according to the actual navigation scenario.

[0041] As an example, navigation information includes, but is not limited to, maps, obstacle information, reference motion trajectories, historical navigation information, and robot motion attribute information. The map can include a map of the robot's environment. The obstacle information can include relevant information about various obstacles present in the robot's environment (such as the obstacle's location, size, and other attribute information).

[0042] The motion trajectory can include reference lines used to guide the robot's movement. Historical navigation information can include various information related to the robot's historical navigation (such as historical motion trajectories, navigation failure records, etc.). The robot's motion attributes can include various attributes that affect or control the robot's motion (such as the robot's own weight, the material of the surface on which the robot is moving, the range of rotation of each joint of the robot, the robot's current pose and speed, etc.).

[0043] Specifically, various navigation information can be determined in different ways depending on the actual situation. For example, maps can be obtained directly from existing maps and / or manually drawn maps. Similarly, obstacle information can be determined using various obstacle detection algorithms.

[0044] As an example, reference trajectories can be flexibly determined in various ways according to actual needs and application scenarios. For instance, a reference path can be a continuous, smooth curve obtained through a series of algorithms (such as path planning algorithms based on the Frenet reference frame, and map information such as road markings analyzed using nonlinear Kalman filters, etc.), closely resembling actual road conditions and thus providing more accurate guidance for the robot's movement. In some cases, reference paths can also be predefined by relevant technicians or the robot's controller. The number of reference paths can be one or more.

[0045] It should be noted that the process of determining navigation information can be performed by the aforementioned entity or by other external devices.

[0046] After obtaining the current navigation information, the aforementioned execution entity can determine the target observation point using various methods. The observation point can include a spatial point to be observed during the robot's movement. The target observation point can be determined based on actual needs.

[0047] As an example, the target observation point can be determined by the following steps: using various target detection algorithms to analyze the current navigation information to identify the spatial points that hinder the robot's movement, as well as the degree of obstruction of each spatial point to the robot's movement, so that at least one spatial point can be selected as the target observation point in order of the degree of obstruction from large to small.

[0048] Step 202: For each candidate motion trajectory in the acquired candidate motion trajectory group, determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point.

[0049] In some embodiments, the aforementioned executing entity may first obtain a set of candidate motion trajectories from its local device and / or other external devices connected in communication with it. The set of candidate motion trajectories may consist of at least one candidate motion trajectory. The candidate motion trajectory may include feasible motion trajectories of the robot in the current environment.

[0050] Specifically, candidate motion trajectories can be flexibly determined in various ways according to actual application needs and scenarios. For example, all feasible motion trajectories can be analyzed based on navigation information such as maps and obstacle information as candidate motion trajectories. Another example is the use of various sampling-based path planning algorithms (such as probabilistic route maps, fast random expansion trees, etc.) to construct paths as candidate motion trajectories.

[0051] It should be noted that the process of determining the candidate motion trajectory can be performed by the aforementioned execution entity or by other external devices.

[0052] After obtaining the candidate motion trajectory group, for each candidate motion trajectory in the candidate motion trajectory group, the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point can be determined as the line-of-sight information corresponding to the candidate motion trajectory.

[0053] The line of sight between the trajectory point and the target observation point can include an imaginary straight line formed between them (such as the projected line of sight from the trajectory point to the target observation point). The line of sight information can represent various information used to describe the line of sight between the trajectory point and the target observation point (such as whether there is occlusion, the type of occlusion, the length and angle of the line of sight, the safety risks that occlusion may pose to the robot, etc.), and can be flexibly set according to the actual application requirements.

[0054] Specifically, the line-of-sight information between the trajectory point and the target observation point can be determined flexibly using various methods. For example, it can be determined by analyzing environmental maps. Alternatively, it can be evaluated through actual measurements.

[0055] Correspondingly, the line-of-sight information corresponding to each candidate motion trajectory can also be determined flexibly in various ways according to actual application needs. For example, the line-of-sight information corresponding to each trajectory point on each candidate motion trajectory can be processed by weighted averaging, and the processing result can be used as the line-of-sight information corresponding to that candidate motion trajectory.

[0056] For example, we can first perform statistical analysis on the line-of-sight information corresponding to each trajectory point on each candidate motion trajectory, and then use the statistical analysis results as the line-of-sight information corresponding to the candidate motion trajectory.

[0057] Optionally, the trajectory points on each candidate motion trajectory may include at least two pre-aiming points on that candidate motion trajectory. These pre-aiming points can be pre-set trajectory points. For example, some trajectory points obtained based on sampling can be used as pre-aiming points.

[0058] In some cases, the aiming point can also be determined based on the target observation point. For example, the aiming point can include a trajectory point that meets preset requirements. As an example, the aiming point can be a trajectory point whose distance from the target observation point meets a preset distance requirement. As yet another example, the aiming point can also be a trajectory point whose line of sight to the target observation point is not obstructed or not completely obstructed.

[0059] Setting up pre-aiming points, such as those that can continuously observe the target observation point, helps ensure that the generated target trajectory can also continuously observe the target observation point, thereby improving the efficiency of target trajectory generation and navigation effect.

[0060] Step 203: Select candidate motion trajectories from the candidate motion trajectory group based on the determined line-of-sight information.

[0061] In some embodiments, after obtaining the line-of-sight information corresponding to each candidate motion trajectory in the candidate motion trajectory group, various methods can be flexibly adopted to select candidate motion trajectories from the candidate motion trajectory group according to actual application requirements.

[0062] For example, a correspondence between line-of-sight information and candidate motion trajectories can be pre-set. After obtaining the line-of-sight information corresponding to each candidate motion trajectory, the corresponding candidate motion trajectory can be determined by querying the above correspondence, and the queried candidate motion trajectory can be selected from the candidate motion trajectory group.

[0063] For example, after obtaining the line-of-sight information corresponding to each candidate motion trajectory, the following steps can be used to select candidate motion trajectories from the candidate motion trajectory group: Step 1: Determine the cost information corresponding to each candidate motion trajectory based on the line-of-sight information.

[0064] In this step, the cost information corresponding to each candidate motion trajectory can represent the cost required to ensure that the robot's motion on that candidate trajectory conforms to a preset motion target. The preset motion target can be flexibly set by relevant technicians or the robot's controller according to actual application needs. For example, the preset motion target may include a motion safety level higher than a preset safety threshold, a motion smoothness level higher than a preset smoothness threshold, a motion stability level higher than a preset stability threshold, and so on.

[0065] Specifically, a conversion expression between line-of-sight information and cost information can be pre-designed. Then, after obtaining the line-of-sight information corresponding to each candidate motion trajectory, the corresponding cost information can be obtained using the above conversion expression.

[0066] Step 2: Based on the cost information corresponding to each candidate motion trajectory, select candidate motion trajectories that meet the preset motion constraints from the candidate motion trajectory group.

[0067] In this step, the preset motion constraints can be flexibly set by relevant technicians or robot controllers according to actual application requirements. For example, preset motion constraints may include minimizing the cost indicated by cost information or ensuring that the cost falls within a preset cost range, etc.

[0068] As an example, if the preset cost target is to be minimized, then the candidate motion trajectory with the minimum cost indicated by the corresponding cost information can be selected from the candidate motion trajectory group.

[0069] Step 204: Generate the target motion trajectory based on the selected candidate motion trajectories, and perform navigation based on the target motion trajectory.

[0070] In some embodiments, after selecting candidate motion trajectories, various methods can be used to generate the target motion trajectory according to actual application requirements. For example, the selected candidate motion trajectory can be directly determined as the target motion trajectory.

[0071] For example, various adjustment methods (such as various trajectory optimization methods) can be used to adjust the selected candidate motion trajectory to optimize the selected candidate motion trajectory, thereby obtaining the adjusted candidate motion trajectory, and then the adjusted candidate motion trajectory can be determined as the target motion trajectory.

[0072] After determining the target trajectory, the robot can be further controlled to move according to the determined target trajectory in order to achieve robot navigation.

[0073] By determining the target observation point and selecting candidate motion trajectories for navigation based on the line-of-sight information of the trajectory points on each candidate motion trajectory toward the target observation point, the robot can proactively select motion trajectories with good line-of-sight for key observation points in advance, thereby achieving proactive safety assurance for the robot.

[0074] In some optional implementations of certain embodiments, the target observation point can be determined based on the current navigation information through the following steps: Step 1: Determine the candidate point set based on the current navigation information.

[0075] In this step, the candidate point set may include at least one of the following candidate points: dynamic points, probe points, and preset guide points.

[0076] Dynamic points can include the locations of obstacles. Obstacles can refer to various obstacles in the robot's environment. For static obstacles, their locations can be directly used as dynamic points. For dynamic obstacles, their movement positions can be predicted based on their velocity, acceleration, and predicted collision time with the robot, and these predicted movement positions can be used as dynamic points. The velocity, acceleration, and predicted collision time of dynamic obstacles can be determined using various image analysis or collision testing methods.

[0077] Detection points can include blind zone boundary points. Specifically, blind zone boundary points can refer to the boundary points of the robot's field of vision blind areas. Various advanced detection technologies can be used to identify known and unknown areas (i.e., blind areas) in the robot's environment, allowing the intersection of these areas to be used as blind zone boundary points.

[0078] Preset guide points can include points pre-set by relevant technical personnel. For example, a preset guide point can be a point on the reference path located a preset look-ahead distance in front of the robot. The preset look-ahead distance can be determined based on historical experience values ​​or simulation experiments.

[0079] Generally, after acquiring the current navigation information, a candidate point set can be determined using corresponding methods based on the required candidate point type. For example, dynamic points can be determined based on surrounding obstacles. Another example is determining detection points based on identified blind spots. Yet another example is setting guide points according to preset look-ahead distances and reference paths, and so on.

[0080] Step 2: Determine the navigation value information of each candidate point in the candidate point set.

[0081] In this step, the navigation value information of a candidate point represents its value for robot navigation. Generally, different methods can be used to determine the navigation value information for different types of candidate points.

[0082] As an example, for dynamic points, various collision algorithms can be used to evaluate their navigation value information. For instance, the probability of a collision between the obstacle corresponding to the dynamic point and the robot can be estimated to represent its navigation value information. Another example is that metrics such as Time to Collision (TTC) can be predicted based on the relative speed between the obstacle corresponding to the dynamic point and the robot to represent its navigation value information.

[0083] For a detection point, its navigation value can be represented by the distance between it and the robot, or by the reduction of environmental information entropy at its boundary.

[0084] For preset guide points, their navigation value information can be set by relevant technical personnel based on their relative position to the robot.

[0085] Step 3: Based on the determined navigation value information, select candidate points from the candidate point set as target observation points.

[0086] In this step, after obtaining the navigation value information of each candidate point in the candidate point set, various methods can be flexibly used to select candidate points from the candidate point set as target observation points.

[0087] For example, candidate points can be selected from the candidate point set as target observation points according to the order of navigation value represented by the corresponding navigation value information from high to low.

[0088] For example, different weights can be set for the navigation value information corresponding to different types of candidate points. In this case, the navigation value represented by the navigation value information of each candidate point can be evaluated by combining the weights, and then the candidate point with the highest corresponding navigation value can be selected from the candidate point set as the target observation point.

[0089] As an example, the weight of a preset guide point can be set to be greater than that of a dynamic point and a probe point, so that the robot can actively observe the global planned path ahead as much as possible.

[0090] By analyzing the navigation value of obstacles, blind zone boundaries, and preset guide points in the environment, observation points that have a key impact on robot navigation are selected as the main visual focus to generate motion trajectories. This guides the robot to actively observe the spatial points that require the most visual attention or have scarce information during movement, thereby planning motion trajectories with better visibility at the observation points in advance.

[0091] In some optional implementations of certain embodiments, for each candidate motion trajectory in the candidate motion trajectory group, at least one of the following can be determined first: trajectory deviation information, collision risk information, and tracking feasibility information corresponding to the candidate motion trajectory. Then, a candidate motion trajectory can be selected from the candidate motion trajectory group based on at least one of the following: trajectory deviation information, collision risk information, and tracking feasibility information, and line-of-sight information.

[0092] Trajectory deviation information characterizes the difference between a candidate trajectory and a reference trajectory. Generally, various trajectory deviation algorithms can be used to determine the deviation between the candidate trajectory and the reference trajectory to represent the trajectory deviation information.

[0093] Collision risk information can characterize the collision risk of a robot's motion on a candidate trajectory. Generally, collision risk information can be determined based on the collision risk between the robot and obstacles in its environment, or it can be determined by combining the potential field value of the map corresponding to the robot's environment.

[0094] Tracking feasibility information characterizes the feasibility of a robot's motion tracking a target observation point along a candidate trajectory. Generally, various target tracking algorithms can be used to determine this information. When the robot has components such as support devices, the difference between the theoretical angular velocity required for continuous tracking of the target observation point by the support device and its maximum angular velocity can also be used to represent tracking feasibility. Typically, tracking may fail when the theoretical angular velocity of the support device is close to or exceeds its maximum angular velocity, indicating low tracking feasibility. Therefore, explicitly considering the dynamic constraints of the support device (such as angular velocity and maximum angular velocity) can improve the smoothness and stability of the tracking trajectory.

[0095] In practical applications, candidate motion trajectories can be flexibly selected from the candidate motion trajectory group based on at least one of the trajectory deviation information, collision risk information, tracking feasibility information, and line-of-sight information corresponding to each candidate motion trajectory.

[0096] For example, constraints on trajectory deviation information, collision risk information, tracking feasibility information, and line-of-sight information can be preset. Then, candidate motion trajectories can be selected that satisfy the constraints on trajectory deviation information, collision risk information, tracking feasibility information, and line-of-sight information.

[0097] For example, for each candidate motion trajectory, the trajectory deviation information, collision risk information, tracking feasibility information and line of sight information corresponding to the candidate motion trajectory can be weighted and fused. At this time, the candidate motion trajectory can be selected from the candidate motion trajectory group according to the weighted fusion result corresponding to each candidate motion trajectory.

[0098] Based on a clear field of view of the target observation point, and considering constraints such as trajectory deviation, collision risk, and tracking feasibility, we further ensure that a motion trajectory with small deviation, low collision risk, and sustainable tracking of the target observation point can be selected, thereby further improving the safety and smoothness of robot navigation.

[0099] Further reference Figure 3 The diagram illustrates flow 300 of some other embodiments of the navigation method for a robot according to the present disclosure. The navigation method for a robot includes the following steps: Step 301: Determine the target observation point based on the current navigation information.

[0100] Step 302: For each candidate motion trajectory in the acquired candidate motion trajectory group, determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point.

[0101] Step 303: Select candidate motion trajectories from the candidate motion trajectory group based on the determined line-of-sight information.

[0102] Step 304: Generate the target motion trajectory based on the selected candidate motion trajectories.

[0103] In some embodiments, the specific implementation of steps 301-304 and the resulting technical effects can be found in [reference needed]. Figure 2 Steps 201-204 in the corresponding embodiments will not be repeated here.

[0104] Step 305: Based on the obtained robot motion parameters and target motion trajectory, generate a tracking trajectory for tracking the target observation point.

[0105] In some embodiments, the aforementioned executing entity may first obtain the robot's motion parameters from local or other external devices. These motion parameters may include various kinematic parameters that support the robot's movement. For example, motion parameters may include the robot's position, velocity, acceleration, etc. Alternatively, motion parameters may include the kinematic parameters of various components or devices associated with the robot.

[0106] As an example, motion parameters can include the rotation angles of robot joints. Motion parameters can also include the rotational range, rotational speed, angular velocity, angular acceleration, etc., of the support equipment mounted on the robot. The support equipment can be various devices (such as various stabilizers, gimbals, etc.) used to support components such as cameras and sensors.

[0107] It should be noted that the robot's motion parameters can be collected by the robot in real time or determined by other external devices.

[0108] The aforementioned tracking trajectory can be used to track target observation points. That is, the tracking trajectory is used to observe or monitor target observation points. Specifically, after obtaining the robot's motion parameters and the target motion trajectory, various methods can be used to generate the tracking trajectory according to the actual application scenario. For example, the tracking trajectory can be generated based on various target tracking algorithms.

[0109] For example, taking the tracking of a target observation point by a robot using a visual sensor and a support device (such as a gimbal) to support the visual sensor as an example, the tracking trajectory can be generated through the following steps: Step 1: For each time step on the target's trajectory, determine the theoretical angle required for the support device to orient the robot's vision sensor toward the target observation point.

[0110] In this step, the theoretical rotation angle represents the angle that the support device theoretically needs to rotate to ensure that the robot's vision sensors are aligned with the target observation point. Specifically, various estimation methods can be used to determine the theoretical rotation angle.

[0111] For example, the pose of the support equipment and the position of the target observation point can be pre-defined, along with the correspondence between these and the theoretical rotation angle of the support equipment. Then, based on the pose of the support equipment at each time step and the position of the target observation point, the theoretical rotation angle of the support equipment at each time step can be determined by querying this correspondence. Alternatively, the theoretical rotation angle of the support equipment at each time step can be determined through simulation.

[0112] Step 2: Generate the tracking trajectory based on the theoretical rotation angle and the robot's motion parameters corresponding to each time step.

[0113] Specifically, after obtaining the theoretical turning angles corresponding to each time step on the target motion trajectory, various methods can be flexibly used to generate the tracking trajectory by combining various motion parameters of the robot.

[0114] As an example, a pre-defined correspondence between the robot's motion parameters and the angular intervals of the support equipment can be established. Then, the angular intervals matching the robot's motion parameters at each time step can be queried, and it can be determined whether the queried angular intervals include the theoretical angular interval. If they do, the theoretical angular interval corresponding to that time step can be used as the target angular interval for that time step; if not, the angular interval closest to the corresponding theoretical angular interval can be selected as the target angular interval for that time step. Furthermore, the angular sequence formed by the target angular intervals corresponding to each time step in chronological order can be used as the tracking trajectory, enabling the robot to control the support equipment to move according to the angular intervals indicated by the tracking trajectory to achieve continuous tracking of the target observation point.

[0115] As another example, relevant technicians can pre-construct a loss function to determine the target rotation angle corresponding to each time step based on the actual application scenario. This loss function can use various information such as the theoretical rotation angle corresponding to each time step and the robot's motion parameters to construct constraints. Then, the loss function can be solved based on the theoretical rotation angle corresponding to each current time step and the robot's motion parameters to determine the target rotation angle corresponding to each time step. Furthermore, the rotation angle sequence formed by the target rotation angles corresponding to each time step in chronological order can be used as the tracking trajectory.

[0116] The constraints can be flexibly set by relevant technical personnel according to actual application needs. For example, constraints may include minimizing the angle difference corresponding to each time step. The angle difference corresponding to each time step can represent the difference between the theoretical angle and the target angle at that time step. Specifically, the angle difference corresponding to each time step can be represented in various ways, such as using the sum of the squares of the norms of the angle differences corresponding to each time step.

[0117] For example, constraints can include maximizing motion smoothness. Motion smoothness can be represented by constraining the angular velocity and / or angular acceleration of the supporting equipment, such as by representing the motion smoothness using a weighted sum of the smoothness constraints of the target rotation angle corresponding to each time step. In this case, the smoothness constraint of each target rotation angle can be represented by the sum of the squares of the norms of the first derivative and the second derivative of that target rotation angle.

[0118] It should be noted that the number of constraints can be arbitrary. Each constraint can also consist of one or more constraints. For example, constraints can include both minimizing the angle difference at each time step and maximizing motion smoothness, as mentioned above. Alternatively, a weighted sum of the angle difference and motion smoothness at each time step can be determined, and minimizing this weighted sum can be used as a constraint.

[0119] For example, the constraint condition can be that the target rotation angle at each time step is no greater than the maximum rotational speed of the robot's support equipment.

[0120] Step 306: Navigate based on the target's movement trajectory and the tracking trajectory.

[0121] After obtaining the robot's target motion trajectory and tracking trajectory, the robot's walking device can be controlled to walk according to the target motion trajectory, while the robot's support equipment can be controlled to rotate according to the tracking trajectory, thereby realizing the robot's navigation.

[0122] While determining the motion trajectory with good line of sight to the target observation point, the tracking trajectory of the robot's visual sensor support equipment to the target observation point is generated. This allows the robot to continuously observe the line of sight information to the target observation point during movement, thereby adjusting the target observation point in a timely manner to plan new motion trajectories and further ensuring the safety and smoothness of the robot's navigation process.

[0123] In some alternative implementations of some embodiments, in response to the robot detecting a trajectory update event while moving along the target trajectory, a new target trajectory can be generated based on the current navigation information at the update time, the target observation point, the candidate trajectory group, and the line-of-sight information, and navigation can be performed based on the new target trajectory.

[0124] The trajectory update events can include various events that trigger the robot to update its motion trajectory, and the specific events can be set according to the actual application scenario and requirements. For example, trajectory update events include, but are not limited to: detecting that the robot's line of sight is obstructed, detecting a new security threat target, and the current target observation point has been sufficiently observed, etc.

[0125] Among these, "safety threat targets" can represent targets whose threat to the robot's movement safety has a higher priority than a preset priority. Typically, safety threat targets are set by relevant technical personnel based on the robot's movement scenario. For example, in the scenario of a home robot, active family members can be considered safety threat targets.

[0126] The conditions for sufficient observation of a target observation point can also be preset by relevant technical personnel according to time and application scenario. For example, when the line-of-sight information corresponding to the selected candidate motion trajectory meets the preset conditions, it can be considered that the target observation point has been sufficiently observed.

[0127] The update time can include the time when the trajectory update event was detected. Specifically, if the robot detects a trajectory update event while moving along the target trajectory, the time when the trajectory update event was detected can be recorded as the update time.

[0128] When a trajectory update event is detected, a new target motion trajectory can typically be generated according to the method described in the above embodiments, based on the navigation information, target observation point, candidate motion trajectory group, and line-of-sight information redefined at the update time. The robot can then be navigated according to the new target motion trajectory. by Figure 4 As an example, flowcharts of further embodiments of the navigation method for robots according to this disclosure are shown. Figure 4 As shown, the robot first determines the target observation point based on the current map, obstacle information, and reference path. Then, it generates the target motion trajectory based on the target observation point and candidate motion trajectories. Next, it generates a tracking trajectory based on the target observation point and the target motion trajectory. The robot then uses the target motion trajectory and the tracking trajectory to navigate and control its chassis and gimbal, respectively. During navigation and control according to the current target motion trajectory and tracking trajectory, the robot uses motion and environmental perception to determine whether a trajectory update is triggered, such as detecting an unexpected obstruction of its line of sight, detecting a new high-priority threat, or determining that the current target observation point has been sufficiently observed. If no trajectory update is triggered, navigation and control can continue according to the current target motion trajectory and tracking trajectory. If a trajectory update is triggered, the above process can be re-executed to begin a new navigation cycle, thus achieving adaptation to the dynamic environment.

[0129] from Figure 4 It can be seen that the robot's navigation process can be decoupled and sequentially executed into three logically progressive and interdependent modular steps: determining the target observation point, generating the target motion trajectory, and generating the tracking trajectory. Based on this, the robot's navigation decision-making logic aligns with human decision-making logic, and, while ensuring computational feasibility, helps achieve a high degree of coordination between robot motion, perception, and active guidance.

[0130] Further reference Figure 5 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a navigation device for a robot, these device embodiments being similar to... Figure 2Corresponding to the method embodiments shown, this navigation device for robots can be specifically applied to various robots.

[0131] like Figure 5 As shown, a navigation device 500 for a robot includes a determining unit 501, a selecting unit 502, and a navigation unit 503. The determining unit 501 is configured to: determine a target observation point based on current navigation information; and for each candidate motion trajectory in a group of acquired candidate motion trajectories, determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point. The selecting unit 502 is configured to: select a candidate motion trajectory from the group of candidate motion trajectories based on the determined line-of-sight information. The navigation unit 503 is configured to: generate a target motion trajectory based on the selected candidate motion trajectory, and perform navigation based on the target motion trajectory.

[0132] In some optional implementations of some embodiments, the determining unit 501 may be further configured to: determine a candidate point set based on current navigation information, wherein the candidate point set includes at least one of the following candidate points: dynamic points, detection points, and preset guide points, wherein dynamic points include the location of dynamic obstacles, and detection points include blind zone boundary points; determine navigation value information for each candidate point in the candidate point set; and select a candidate point from the candidate point set as a target observation point based on the determined navigation value information.

[0133] In some optional implementations of some embodiments, the determining unit 501 may be further configured to: for each candidate motion trajectory in the candidate motion trajectory group, determine at least one of trajectory deviation information, collision risk information, and tracking feasibility information corresponding to the candidate motion trajectory, wherein the trajectory deviation information characterizes the difference between the candidate motion trajectory and the reference motion trajectory, the collision risk information characterizes the collision risk of the robot motion on the candidate motion trajectory, and the tracking feasibility information characterizes the feasibility of the robot motion tracking the target observation point on the candidate motion trajectory; and the selecting unit 502 may be further configured to: select a candidate motion trajectory from the candidate motion trajectory group based on at least one of the trajectory deviation information, collision risk information, and tracking feasibility information and the line of sight information.

[0134] In some optional implementations of some embodiments, the navigation unit 503 may be further configured to: generate a tracking trajectory for tracking the target observation point based on the acquired motion parameters of the robot and the target motion trajectory; and perform navigation based on the target motion trajectory and the tracking trajectory.

[0135] In some optional implementations of certain embodiments, the apparatus 500 further includes an update unit (not shown in the figure). The update unit can be configured to: in response to detecting a trajectory update event during movement along a target trajectory, generate a new target trajectory based on current navigation information at the update time, the target observation point, candidate trajectory groups, and line-of-sight information, and perform navigation based on the new target trajectory.

[0136] It is understandable that the units described in the navigation device 500 for robots are related to the reference... Figure 2 The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the navigation device 500 for the robot and the units contained therein, and will not be repeated here.

[0137] The following is for reference. Figure 6 It illustrates robots suitable for implementing some embodiments of this disclosure (e.g., Figure 1 The structural diagram of robot 101)600. Figure 6 The robot shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0138] like Figure 6 As shown, the robot 600 may include a vision sensor 601, a processing module 602, and a control module 603. The vision sensor 601 can be various devices capable of acquiring visual information, such as a camera, image sensor, light projector, etc. The processing module 602 can be various devices capable of processing image and other information (such as a microprocessor). The control module 603 can be various devices capable of controlling the movement of various robot components (such as a microcontroller). In some cases, the processing module 602 and the control module 603 may be integrated into a single chip.

[0139] The vision sensor 601 can acquire current navigation information. The processing module 602 can determine the target observation point based on the current navigation information; for each candidate motion trajectory in the acquired candidate motion trajectory group, it determines the line-of-sight information of the trajectory points on the candidate motion trajectory towards the target observation point; based on the determined line-of-sight information, it selects a candidate motion trajectory from the candidate motion trajectory group; and it generates the target motion trajectory based on the selected candidate motion trajectory. The control module 603 can navigate the robot according to the target motion trajectory.

[0140] Typically, robot 600 may also include support equipment (not shown in the figure). The support equipment can be various devices capable of supporting vision sensors (such as a gimbal). Processing module 602 can also acquire the robot's motion parameters and, using the motion parameters and the target motion trajectory, generate a tracking trajectory for tracking the target observation point. Control module 603 can also navigate the robot based on the target motion trajectory and the tracking trajectory.

[0141] In addition, depending on actual needs, the robot may also include, but is not limited to, the following components or modules (not shown in the figure): storage modules (such as various memory devices), input / output components (such as microphones, speakers, displays, etc.), and communication components (such as Bluetooth modules, near-field communication modules, etc.).

[0142] Although Figure 6 Robot 600 with various parts or modules is shown; however, it should be understood that implementation or possession of all shown parts or modules is not required. More or fewer parts or modules may be implemented alternatively. Figure 6 Each box shown can represent a component or module, or multiple components or modules as needed.

[0143] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be installed. When the computer program is executed by robot 600, it performs the functions defined in the methods of some embodiments of this disclosure.

[0144] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0145] The aforementioned computer-readable medium may be included in the robot or may exist independently without being assembled into the robot. The aforementioned computer-readable medium carries one or more programs that, when executed by the robot, cause the robot to: determine a target observation point based on current navigation information; for each candidate motion trajectory in the acquired candidate motion trajectory group, determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point; select a candidate motion trajectory from the candidate motion trajectory group based on the determined line-of-sight information; generate a target motion trajectory based on the selected candidate motion trajectory; and navigate based on the target motion trajectory.

[0146] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of robots, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0148] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a determining unit, a selecting unit, and a navigation unit. The names of these units do not necessarily limit the specific unit; for example, a determining unit may also be described as "a unit that determines a target observation point based on current navigation information."

[0149] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0150] Some embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements any of the above-described navigation methods for a robot.

[0151] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A navigation method for a robot, comprising: Determine the target observation point based on the current navigation information; For each candidate motion trajectory in the acquired candidate motion trajectory group, determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point; Based on the determined line-of-sight information, a candidate motion trajectory is selected from the candidate motion trajectory group; Generate a target motion trajectory based on the selected candidate motion trajectories, and perform navigation based on the target motion trajectory.

2. The method according to claim 1, wherein, The step of determining the target observation point based on the current navigation information includes: Based on the current navigation information, a candidate point set is determined, wherein the candidate point set includes at least one of the following candidate points: dynamic points, detection points, and preset guide points, wherein the dynamic points include the locations of dynamic obstacles, and the detection points include blind zone boundary points; Determine the navigation value information of each candidate point in the candidate point set; Based on the determined navigation value information, candidate points are selected from the candidate point set as target observation points.

3. The method according to claim 1, wherein, The method further includes: For each candidate motion trajectory in the candidate motion trajectory group, at least one of the following is determined: trajectory deviation information, collision risk information, and tracking feasibility information. The trajectory deviation information characterizes the difference between the candidate motion trajectory and the reference motion trajectory; the collision risk information characterizes the collision risk of the robot's motion on the candidate motion trajectory; and the tracking feasibility information characterizes the feasibility of the robot's motion tracking the target observation point on the candidate motion trajectory. The step of selecting candidate motion trajectories from the candidate motion trajectory group based on the determined line-of-sight information includes: Based on at least one of the trajectory deviation information, collision risk information, and tracking feasibility information, and the line-of-sight information, a candidate motion trajectory is selected from the candidate motion trajectory group.

4. The method according to any one of claims 1-3, wherein, The method further includes: Based on the acquired motion parameters of the robot and the target motion trajectory, a tracking trajectory for tracking the target observation point is generated; and The navigation based on the target's motion trajectory includes: Navigation is performed based on the target's movement trajectory and the tracking trajectory.

5. The method according to any one of claims 1-3, wherein, The method further includes: In response to detecting a trajectory update event during movement along the target trajectory, a new target trajectory is generated based on the current navigation information, target observation point, candidate trajectory group, and line-of-sight information at the update time, and navigation is performed based on the new target trajectory, wherein the update time includes the time when the trajectory update event was detected.

6. A navigation device for a robot, comprising: The determining unit is configured to determine the target observation point based on the current navigation information; The determining unit is further configured to determine the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point for each candidate motion trajectory in the acquired candidate motion trajectory group; The selection unit is configured to select a candidate motion trajectory from the candidate motion trajectory group based on determined line-of-sight information; The navigation unit is configured to generate a target motion trajectory based on selected candidate motion trajectories, and to navigate based on the target motion trajectory.

7. A robot, comprising a vision sensor, a processing module, and a control module; The processing module uses the visual sensor to determine the target observation point based on the current navigation information; for each candidate motion trajectory in the acquired candidate motion trajectory group, it determines the line-of-sight information of the trajectory points on the candidate motion trajectory toward the target observation point; Based on the determined line-of-sight information, a candidate motion trajectory is selected from the candidate motion trajectory group; Generate the target motion trajectory based on the selected candidate motion trajectories; The control module navigates according to the target's motion trajectory.

8. The robot according to claim 7, wherein, The robot also includes a support device for supporting the vision sensor; and The processing module generates a tracking trajectory for tracking the target observation point based on the acquired motion parameters of the robot and the target motion trajectory. The control module performs navigation based on the target's motion trajectory and the tracking trajectory.

9. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-5.

10. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-5.

Images