Unmanned aerial vehicle intelligent obstacle avoidance method and system for unknown environment exploration

Abstract

The invention relates to the field of unmanned aerial vehicle detection, in particular to an unmanned aerial vehicle intelligent obstacle avoidance method and system for unknown environment exploration, and the method comprises the steps: defining a safety channel when an unmanned aerial vehicle flies according to the configuration parameters of the unmanned aerial vehicle, then measuring the detection distance of the unmanned aerial vehicle relative to an indoor obstacle in real time, and then calculating the obstacle avoidance of the unmanned aerial vehicle according to the detection distance. And judging whether the indoor obstacle enters the safety channel, if so, calculating a safety angle of the unmanned aerial vehicle relative to the indoor obstacle, and finally sending an instruction of advancing towards the safety angle to the unmanned aerial vehicle. Compared with the prior art, the unmanned aerial vehicle can safely pass in an unknown indoor environment.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) detection. More specifically, the present invention relates to an intelligent obstacle avoidance method and system for UAVs used in unknown environment exploration. Background Art

[0002] A UAV is an aircraft controlled by a remote control or an autonomous system, and its types include fixed-wing, multi-rotor and hybrid UAVs, which are widely used in fields such as reconnaissance, agricultural monitoring, logistics distribution, terrain survey and rescue. When a UAV is flying to find a path, it mainly locates through GNSS (Global Navigation Satellite System). However, GNSS has high requirements for the quality of signal reception. In indoor, bridge, tunnel and underground environments, the positioning signals sent by navigation satellites are difficult to reach the receiving antenna of the UAV, making it impossible for the UAV to locate in the above environments.

[0003] To solve the above problems, in the prior art, a SLAM technology is proposed for indoor positioning of UAVs. It mainly relies on visual cameras, lidar and database information to collect and analyze information on known indoor environments, and constructs a path map based on the analysis results of the data to achieve path navigation of the UAV. However, the above technology is a data analysis and map drawing of known indoor environments, mainly used for indoor inspections. When exploring unknown environments (such as real-time tunneling mine shafts, tunnels or outdoor karst caves), database information cannot be obtained, resulting in the inability to generate a path map, and further making it impossible for the UAV to safely explore in unknown environments. Summary of the Invention

[0004] To solve the technical problem that UAVs cannot safely explore in unknown environments, the present invention provides solutions in the following aspects.

[0005] In a first aspect, the present invention provides an intelligent obstacle avoidance method for UAVs used in unknown environment exploration, including:

[0006] Defining a safety channel when the UAV is flying according to the configuration parameters of the UAV;

[0007] Real-time measuring the detection distance of the UAV relative to indoor obstacles;

[0008] Judging whether the indoor obstacles enter the safety channel according to the detection distance. If so, calculating the safety angle of the UAV relative to the indoor obstacles;

[0009] Sending an instruction to the UAV to move forward at the safety angle.

[0010] Beneficial effects: The method of the present invention customizes a safety channel according to the configuration parameters of the drone, then calculates the safety angle using the safety channel and the detection distance, and finally sends a forward instruction to the drone according to the safety angle, enabling the drone to safely pass through an unknown indoor environment.

[0011] Preferably, calculating the safety angle of the drone relative to indoor obstacles specifically includes:

[0012] Taking the forward direction of the drone as the 0° direction;

[0013] From the upper limit of the preset detection angle to the 0° direction, continuously detect multiple interval distances of the drone relative to indoor obstacles counterclockwise, calculate the rotation angle corresponding to each interval distance, and take the rotation angle corresponding to the interval distance with the minimum length as the minimum angle;

[0014] Starting from the minimum angle as the starting angle, continuously judge counterclockwise whether there is a safety channel without indoor obstacles entering, and take the rotation angle corresponding to the first safety channel that meets the conditions as the safety angle.

[0015] Beneficial effects: The method of the present invention customizes a set of action logics for searching for the safety angle. Using this action logic and the safety angle discrimination algorithm, the method of the present invention can adapt to an unknown indoor environment and make an optimal safety angle judgment. On the one hand, it can enable the drone to move along the edge of the indoor environment to achieve traversal exploration of the indoor environment. On the other hand, it can ensure that the drone will not collide with the edge of the indoor environment, improving the safety of the drone flying in an unknown indoor environment.

[0016] Preferably, the configuration parameters at least include the safe travel width and the safe distance; the safe travel width is greater than the diagonal length of the top-down projection of the drone; the safety channel is a rectangular channel; according to the configuration parameters of the drone, defining the safety channel during the flight of the drone specifically includes:

[0017] Defining the safe travel width as the width of the safety channel;

[0018] Defining the safe distance as the length of the safety channel;

[0019] Defining the position of the drone to the midpoint of the near side of the safety channel.

[0020] Furthermore, the configuration parameters further include the minimum movement distance of the drone moving forward each time, and the length of the minimum movement distance is less than the safe distance.

[0021] Beneficial effects: In the method of the present invention, the safe travel width (greater than the diagonal length of the top-down projection of the drone) is used as the width of the safe passage to ensure that the drone will not collide with obstacles when traveling or rotating in the safe passage. The method of the present invention also uses the safe distance (greater than the minimum movement distance of the drone) as the length of the safe passage to ensure that the drone always moves in the obstacle-free area during a single forward movement, so as to improve the safety of the drone moving in an unknown indoor environment. Compared with the prior art, the safe passage of the method of the present invention can be adaptively adjusted according to the configuration parameters of the drone and has higher reliability.

[0022] Preferably, after issuing a forward command to the drone, the method of the present invention further includes:

[0023] After the drone responds to the forward command and moves forward, record the current position of the drone as a safe node;

[0024] Set the return route of the drone according to the safe node.

[0025] Beneficial effects: Every time the drone safely moves to a certain position in an unknown indoor environment, the method of the present invention will mark this position as a safe node and use multiple safe nodes to form the return route of the drone, so that the drone can safely return after the exploration is completed.

[0026] Preferably, after setting the return route of the drone according to the safe node, the method further includes:

[0027] Extract two safe nodes from the return route of the drone;

[0028] Judge whether it is possible to safely pass between the two safe nodes. If so, discard other nodes located within the return route of the drone between the two safe nodes.

[0029] Beneficial effects: After setting the return route of the drone, the above method can screen out unnecessary nodes, shorten the actual safe return journey of the drone, and realize the rapid return of the drone.

[0030] Preferably, if the drone meets the preset return conditions during flight, the method of the present invention further includes:

[0031] Issue a command to the drone to return along the return route of the drone.

[0032] In a second aspect, the present invention also provides a drone intelligent obstacle avoidance system for unknown environment exploration, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the drone intelligent obstacle avoidance method for unknown environment exploration described in the first aspect is implemented.

[0033] The beneficial effects of the present invention are as follows:

[0034] (1) Compared with the prior art, the method of the present invention customizes a safety channel according to the configuration parameters of the unmanned aerial vehicle (UAV), then calculates the safety angle by using the safety channel and the detection distance, and finally sends a forward instruction to the UAV according to the safety angle, enabling the UAV to safely pass through an unknown indoor environment.

[0035] (2) Compared with the prior art, the method of the present invention customizes a set of action logics for searching the safety angle. By using the action logics and the safety angle discrimination algorithm, the method of the present invention can adapt to an unknown indoor environment and make an optimal safety angle judgment. On the one hand, it can enable the UAV to move along the edge of the indoor environment to realize the traversal exploration of the indoor environment. On the other hand, it can ensure that the UAV will not collide with the edge of the indoor environment, improving the safety of the UAV flying in an unknown indoor environment.

[0036] (3) Compared with the prior art, the safety channel of the method of the present invention can be adaptively adjusted according to the configuration parameters of the UAV, with higher reliability. Description of the Drawings

[0037] By reading the following detailed description with reference to the drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts, where:

[0038] Figure 1 is a flowchart of the intelligent obstacle avoidance method for a UAV used in the exploration of an unknown environment in the first embodiment of the present invention;

[0039] Figure 2 is a schematic structural diagram of the intelligent obstacle avoidance system for a UAV used in the exploration of an unknown environment in the third embodiment of the present invention. Detailed Embodiments

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present invention.

[0041] The following will describe the detailed embodiments of the present invention in detail with reference to the drawings.

[0042] This embodiment discloses an intelligent obstacle avoidance method and system for a UAV used in the exploration of an unknown environment, which is used to solve the technical problem that the UAV cannot safely explore in an unknown environment.

[0043] Embodiment 1

[0044] As Figure 1 shown, this embodiment discloses an intelligent obstacle avoidance method for an unmanned aerial vehicle (UAV) used for unknown environment exploration, including:

[0045] S10: Define a safety passage for the UAV during flight according to the configuration parameters of the UAV.

[0046] S20: Measure the detection distance of the UAV relative to indoor obstacles in real time.

[0047] S30: According to the detection distance, determine whether the indoor obstacle enters the safety passage. If so, calculate the safety angle of the UAV relative to the indoor obstacle; if not, send a forward command to the UAV.

[0048] S40: Send a command to the UAV to move forward at the safety angle.

[0049] Through the above steps S10 - S40, the method of the present invention first customizes a safety passage according to the configuration parameters of the UAV, and then determines whether the indoor obstacle is located in the safety passage according to the detection distance. If so, calculates the safety angle and makes the UAV move forward at the safety angle. If not, the UAV moves forward normally. Compared with the prior art, the method of the present invention is more adaptable to the indoor environment, enabling the UAV to safely pass through the unknown indoor environment.

[0050] Further, the configuration parameters of the UAV in the above S10 at least include a safe travel width safe_width, a safe distance safe_distance, and a minimum movement distance min_fly_distance. Generally, the safe travel width safe_width should be greater than the length of the diagonal of the top-down projection to prevent the UAV from rotating and hitting obstacles in the safety passage. The minimum movement distance min_fly_distance refers to the minimum distance that the UAV moves at one time.

[0051] Preferably, the above minimum movement distance min_fly_distance is one-half of the safe distance safe_distance.

[0052] It should be noted that the safe passage in S10 above is a preset rectangular passage facing the detector directly. Its width is defined as the safe travel width safe_width, and its length is defined as the safe distance safe_distance. After defining its width and length, the position of the drone also needs to be defined at the midpoint of the near side of the safe passage. If the indoor obstacle is inside the safe passage, it is regarded that the drone cannot fly in the safe passage. More specifically, calculate the distance between the intersections of the drone and the edges of the safe passage in the directions of 0°-90° and 270°-359°. If the distance of the indoor obstacle in a certain direction is less than this distance, it is determined that the obstacle enters the "safe passage", and the drone cannot fly in this direction.

[0053] Through the above technical solution, the above safe passage can be adaptively adjusted according to the configuration parameters of the drone, with higher reliability.

[0054] Further, in this embodiment, the detector alone uses a rotary lidar to achieve the ranging in step S20. The lidar can give the distance of the obstacles in the 360° horizontal direction of the drone in real time (the laser frequency is generally 5Hz or 10Hz). If the front of the drone is 0°, the clockwise rotation interval is 0°-359° (that is, the front is 0°, the right is 90°, the back is 180°, and the left is 270°), a total of 360 data.

[0055] It should be added that the lidar ranging also has a longitudinal detection angle range. Within this detection angle range, the distance of the nearest indoor obstacle within the detection angle range will be returned.

[0056] In other embodiments, a visual ranging device can also be used alone to achieve the ranging in step S20. If a visual test device is used, a light source needs to be additionally configured on the body of the drone.

[0057] Further, before the process of calculating the safe angle of the drone relative to the indoor obstacle in S30 above, the drone continues to fly in a certain direction until the obstacle enters the safe passage, and a forward rotation command is first sent to the drone.

[0058] In this embodiment, the counterclockwise rotation direction is used as the forward rotation direction. In other embodiments, the clockwise rotation direction can be used as the forward rotation direction. The above forward rotation command means to command the drone to rotate counterclockwise by 90°.

[0059] Then, enter the main loop program, which involves the main loop program logic of S30. Specifically:

[0060] S31: Take the direction directly facing the drone as the 0° direction.

[0061] S32: From the upper limit of the preset detection angle to the 0° direction, continuously detect counterclockwise multiple spaced distances of the drone relative to indoor obstacle objects, calculate the rotation angle corresponding to each spaced distance, and use the rotation angle corresponding to the spaced distance with the minimum length as the minimum angle.

[0062] In this embodiment, the upper limit of the detection angle is 135°. The rotary lidar is used to rotate counterclockwise from 135° to 0°, continuously detect the spaced distances, and then calculate the above-mentioned minimum angle based on the processor built in the drone.

[0063] S33: Start judging with the minimum angle as the starting angle, continuously judge counterclockwise whether there is a safe passage without indoor obstacles entering, and use the rotation angle corresponding to the first safe passage that meets the conditions as the safe angle.

[0064] Specifically, start judging counterclockwise from the minimum angle min_angle whether there is a safe passage without obstacles entering. If so, record the angle corresponding to the first safe passage that meets the conditions as safe_angle, which is the next forward direction of the drone.

[0065] Through the above steps S31 - S33, the drone controlled by the method of the present invention can move along the edge of the indoor environment counterclockwise, realizing the traversal of indoor exploration, improving the comprehensiveness of exploration of unknown indoor environments. On the other hand, the above steps can ensure that the drone will not collide with the edge of the indoor environment, improving the safety of the drone flying in the unknown indoor environment.

[0066] In the main loop program, step S40 is specifically:

[0067] S41: Send an instruction to the drone to move forward at the safe angle, so that the drone flies forward the minimum moving distance.

[0068] S42: Wait for the drone to fly to the next node, and return to step S31.

[0069] The above main loop steps S31 - S42 can enable the drone to quickly find the edge points in the unknown environment and fly along the edge points in the unknown environment. When the preset return condition is met, jump out of the above main loop and execute the return program.

[0070] Furthermore, in order to enable the drone to return safely after detection, after the above step S40, the method of the present invention further includes:

[0071] After the drone responds to the forward instruction and moves forward, record the current position of the drone as the safe node.

[0072] Set the return route of the drone according to the safe node.

[0073] In this embodiment, after the drone receives the forward instruction and moves forward by the minimum movement distance min_fly_distance, it is considered to have reached the next node. Every time the drone flies to a node, it saves the position information and obstacle information of that node for calculating the path back to the starting point from that node.

[0074] Furthermore, in order to shorten the actual safe return path of the drone and enable the drone to return quickly, after setting the return path of the drone according to the safe nodes, the method of the present invention further includes:

[0075] Extract two safe nodes from the return path of the drone.

[0076] Determine whether it is possible to pass safely between the two safe nodes. If so, discard the other nodes within the return path of the drone between the two safe nodes.

[0077] Specifically, every time the drone reaches a node, according to the preset segment length, it calculates whether it can safely reach the previous node. If it can reach safely, the nodes between these two nodes are discarded and no longer used for the drone's return.

[0078] It should be added that the method for determining whether it can safely reach the previous node is specifically as follows: Using safe_width as the width and the distance from the current node to the previous node as the length, construct a longer temporary safe passage. According to the distance of the obstacles around the current node, determine whether the obstacles enter the temporary safe passage. If so, the drone can fly safely to the previous safe node and discard the other nodes between the two safe nodes.

[0079] In this embodiment, every time the drone reaches a node, it calculates the safe passage forward, and the nodes that have been discarded are no longer used under normal circumstances. In special cases, the discarded nodes will be enabled. For example, if the obstacles in the original return path of the drone move, at this time the drone needs to rely on all existing nodes (including the discarded nodes), calculate the nearest node that it can fly to safely, and then use this node to recalculate the return path.

[0080] Furthermore, if the drone meets the preset return conditions during flight, the method of the present invention further includes:

[0081] Send an instruction to the drone to return along the return path of the drone.

[0082] Specifically, the above return conditions include:

[0083] (1) The drone has traversed an unknown indoor environment and returned to the starting node.

[0084] (2) The battery power of the drone is lower than the preset value.

[0085] (3) The drone calculates based on the power consumption of the traveled distance and the power consumption of the drone's return route. If the remaining power cannot meet the power consumption of the drone's return route, it has to return.

[0086] Preferably, to prevent the drone from flying too far from the starting point, a maximum distance max_distance can be preset. When it is calculated that the distance of the next node of the drone from the starting point is greater than max_distance, the drone searches for a "safe passage" to return with the tangent direction of the line connecting the node and the starting point as the starting direction, so as to ensure that the distance of the next node from the starting point will not exceed max_distance.

[0087] Embodiment 2

[0088] Based on the realization of obstacle detection, avoidance and path tracking in the horizontal direction in Embodiment 1, this embodiment also discloses an intelligent obstacle avoidance method for drones used in unknown environment exploration, which is used to realize obstacle detection, avoidance and path tracking in the vertical direction. After S20, the method of the present invention further includes:

[0089] S200: Obtain the laser point cloud data at the top or bottom of the drone.

[0090] Specifically, the laser point cloud data at the top or bottom of the drone is obtained by adding a lidar at the top or bottom of the drone.

[0091] S300: Segment the laser point cloud data, and segment it into wall points and non-wall points.

[0092] It should be explained that the initial height of the drone entering the unknown indoor environment can be set according to the actual situation. Generally, if the initial height is set reasonably, only using obstacle detection, avoidance and path tracking in the horizontal direction can achieve the safe passage of the drone in the unknown indoor environment. In special cases, such as underground karst caves, vertically extending mine shafts, etc., the effect achieved by only using the technical solution described in Embodiment 1 is not ideal enough. Therefore, it is necessary to perform point cloud data analysis in the vertical direction. Segmenting the laser point cloud data into wall points and non-wall points can quickly enable the drone to find the top and bottom wall references, maintain the height of the drone relative to the ground, and keep the drone at a safe flight height all the time.

[0093] S400: Cluster the non-wall points, judge the types of the non-wall points after clustering, and record the nodes.

[0094] Specifically, before identifying the non-wall points, cluster multiple consecutive non-wall points, and identify the vertical channels and indoor obstacles according to the clustering results and the preset channel judgment mechanism.

[0095] More specifically, a top safety passage and a bottom safety passage are respectively set at the top and bottom of the drone, and length thresholds of the top safety passage and the bottom safety passage are defined. If the non-wall points after the above clustering exceed the top safety passage or the bottom safety passage, it indicates that there are cavities, i.e., vertical passages, at the top or bottom of the drone. At this time, the drone maintains its initial height and executes steps S30 - S40, while recording the nodes where vertical passages exist. If the non-wall points after the above clustering are within the top safety passage or the bottom safety passage, it indicates that there are indoor obstacles at the top or bottom of the drone. Record the nodes with indoor obstacles and execute steps S30 - S40.

[0096] S500: Calculate the distance between the wall points and the drone. If the distance is lower than the preset safety distance, send a height adjustment instruction to the drone.

[0097] Specifically, in an unknown indoor environment, there is a situation where the wall gradually shrinks. If the drone always flies forward at the initial height, it may scrape against the upper wall or the lower wall in the room, and in severe cases, the propeller of the drone may break, resulting in the drone being unable to continue flying forward. Therefore, before executing steps S30 - S40, the method of this embodiment calculates the distance between the wall points and the drone. If the distance is lower than the preset safety distance, a height adjustment instruction is sent to the drone to increase or decrease the flight height of the drone until the distance is greater than or equal to the safety distance. After the above step S500 is completed, steps S30 - S40 are executed.

[0098] Compared with the prior art, the method of the present invention can be applied to the exploration of almost all kinds of complex unknown indoor environments, and has higher safety and reliability.

[0099] Embodiment III

[0100] As Figure 2 shown, this embodiment discloses a drone intelligent obstacle avoidance system for unknown environment exploration, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the drone intelligent obstacle avoidance method for unknown environment exploration described in Embodiment I or II is implemented.

[0101] In the present invention, the foregoing memory may be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as, resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0102] In the description of this specification, "a plurality of" means at least two, for example, two, three, or more, etc., unless otherwise specifically and clearly defined.

[0103] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in the practice of the present invention.

Claims

1. An intelligent obstacle avoidance method for unmanned aerial vehicles used for exploring unknown environments, characterized in that: include: Defining a safe passage for the UAV to fly according to the configuration parameters of the UAV; Measuring the detection distance of the drone relative to indoor obstacles in real time; According to the detection distance, determine whether the indoor obstacle enters the safety passage, and if so, calculate the safety angle of the drone relative to the indoor obstacle; A command is issued to the drone to move toward the safety angle.

2. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 1 is characterized in that: Calculate the safety angle of the drone relative to the indoor obstacle, specifically: The facing direction of the UAV is taken as the 0° direction; From the preset detection angle upper limit to the 0° direction, the drone is continuously detected in a counterclockwise direction at multiple intervals relative to the indoor obstacle, the rotation angle corresponding to each interval is calculated, and the rotation angle corresponding to the interval with the shortest length is taken as the minimum angle; The judgment is started with the minimum angle as the starting angle, and it is continuously judged in a counterclockwise direction whether there is a safe passage without indoor obstacles, and the rotation angle corresponding to the first safe passage that meets the conditions is taken as the safe angle.

3. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 1, characterized in that: The configuration parameters include at least a safe travel width and a safe distance; the safe travel width is greater than the diagonal length of the top view projection of the UAV; the safe passage is a rectangular passage; according to the configuration parameters of the UAV, the safe passage of the UAV during flight is defined as follows: Define the safe travel width as the width of the safety passage; The safety distance is defined as the length of the safety passage; The position of the drone is defined as the near side midpoint of the safety channel.

4. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 3 is characterized in that: The configuration parameters also include a minimum moving distance of the drone for a single forward movement, and the length of the minimum moving distance is less than the safety distance.

5. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 1, characterized in that: After issuing a forward instruction to the drone, the method further includes: After the UAV responds to the forward instruction and moves forward, the current position of the UAV is recorded as a safety node; According to the safety node, the return route of the drone is set.

6. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 5, characterized in that: After setting the return route of the drone according to the safety node, the method further includes: Extracting two safety nodes in the return route of the UAV; Determine whether two safety nodes can pass safely. If so, abandon other nodes between the two safety nodes within the return route of the drone.

7. The intelligent obstacle avoidance method for unmanned aerial vehicle for exploring unknown environments according to claim 5, characterized in that: If the drone meets the preset return conditions during flight, the method further includes: A command is issued to the drone to return along the drone's return route.

8. An intelligent obstacle avoidance system for unmanned aerial vehicles used for exploring unknown environments, characterized in that: It includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the unmanned aerial vehicle intelligent obstacle avoidance method for unknown environment exploration according to any one of claims 1 to 7 is implemented.

Images