Unmanned garbage sweeper active defense system based on multi-sensor fusion and application

Abstract

The invention discloses an unmanned garbage sweeper active defense system based on multi-sensor fusion and application, and the system is characterized in that a road end monitoring module collects camera perception data in a service area in real time, and sends object movement information in the service area to a fusion module; the vehicle-mounted non-contact sensing module collects laser sensing data, camera sensing data and auxiliary sensing data of the motor sweeper in real time and sends the data to the fusion module; the vehicle-mounted contact sensing module collects contact sensing data of the sweeping vehicle and target garbage in real time and sends the contact sensing data to the fusion module; the fusion module fuses the camera sensing data, the laser sensing data, the auxiliary sensing data and the contact sensing data, outputs obstacle target information and sends the obstacle target information to the decision execution module; and after receiving the obstacle target information, the decision execution module carries out decision judgment and executes corresponding defense actions, so that the garbage sweeper is prevented from colliding with the obstacle.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned garbage sweepers, and in particular to an active defense system for unmanned garbage sweepers based on multi-sensor fusion and its application. Background Art

[0002] As an important part of highways, the construction of smart service areas has become an important direction for the development of the industry as the domestic highway network continues to be built and people's requirements for travel experience increase. Due to the relatively remote locations of highway service areas, poor cleaning working environment, high intensity, and long time span, the traditional sanitation service industry has problems such as difficulty in recruiting, high cost, and low efficiency. It is urgent to apply intelligent cleaning technology to clean service areas.

[0003] Due to the complex environment in the service area, vehicles generally travel at high speeds on the ramps of the service area, which makes collision accidents prone to occur. Secondly, there are many large vehicles such as trucks and hazardous chemical trucks in the service area, and the visibility is limited, so unmanned garbage sweepers are prone to collisions with surrounding vehicles and obstacles. Therefore, how to identify and judge garbage and obstacles in highway service areas, and at the same time guide unmanned garbage sweepers to perform adaptive cleaning, pushing or avoidance is a topic with positive practical significance. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose an active defense system for unmanned garbage sweepers based on multi-sensor fusion and its application which is reliable in implementation and flexible in application.

[0005] In order to achieve the above technical objectives, the technical solution adopted by the present invention is: An active defense system for unmanned garbage sweepers based on multi-sensor fusion is applied to unmanned garbage sweepers in highway service areas. The unmanned garbage sweepers are equipped with a camera perception module, a laser perception module, an auxiliary perception module and a contact perception module for road surface monitoring. The highway service area is equipped with a camera for monitoring public areas, which includes: a road-side monitoring module, a vehicle-mounted non-contact perception module, a vehicle-mounted contact perception module, a fusion module and a decision execution module. The roadside monitoring module is used to collect perception data captured by cameras deployed in the high-speed service area in real time and send it to the fusion module; The on-board non-contact sensing module is used to collect the sensing data obtained by the camera sensing module, laser sensing module, and auxiliary sensing module deployed on the unmanned garbage sweeper in real time, and send the non-contact sensing data to the fusion module; The vehicle-mounted contact sensing module is used to collect the sensing data obtained when the unmanned garbage sweeper contacts the target garbage in real time, and transmit the contact sensing data to the fusion module; The fusion module fuses the service area camera perception data, non-contact perception data, and contact perception data collected by the roadside monitoring module, the non-contact perception module, and the contact perception module according to preset requirements to output obstacle target information, and then sends it to the decision execution module; After receiving the obstacle target information, the decision execution module makes a decision according to the preset requirements, and then executes the corresponding defense action to avoid the garbage sweeper from colliding with the obstacle.

[0006] As a possible implementation mode, further, the road-side monitoring module described in this scheme is communicatively connected with the security monitoring equipment deployed in the service area and the traffic monitoring equipment on the roadside, both of which have cameras and are used to perform real-time monitoring of designated points in the highway service area and upload them to the local device that is communicatively connected thereto, and then use the preset algorithm to perform real-time recognition of the perception data of the scene images of the designated points captured by the security monitoring equipment and the traffic monitoring equipment, and monitor the movement of objects in the designated points in the service area, and then send them to the fusion module deployed in the unmanned garbage sweeper.

[0007] As a possible implementation mode, further, the vehicle-mounted non-contact sensing module described in this scheme is communicatively connected with multiple vehicle-mounted non-contact sensors deployed on the unmanned garbage cleaning vehicle, including laser radar, camera, millimeter-wave radar, ultrasonic radar, and single-line laser radar, wherein the laser radar is one of the components of the laser sensing module, the camera is one of the components of the camera sensing module, and the millimeter-wave radar, ultrasonic radar, and single-line laser radar are all components of the auxiliary sensing module.

[0008] As a possible implementation mode, further, the vehicle-mounted contact sensing module described in this solution is communicatively connected to a vehicle-mounted pressure sensor deployed on the unmanned garbage cleaning vehicle, and the vehicle-mounted pressure sensor is one of the components of the contact sensing module.

[0009] As a possible implementation mode, further, the fusion module of the present solution fuses the non-contact sensing data and contact sensing data collected by the camera sensing module, the laser sensing module, the auxiliary sensing module and the contact sensing module according to preset requirements, and the data sources complement each other, and after multi-level redundancy, finally output obstacle information based on multi-sensor fusion, wherein the obstacle information includes at least one of the position, speed, shape, size, category and orientation of the obstacle, and the obstacle information is sent to the decision execution module; Among them, the non-contact sensing data includes camera sensing data, laser sensing data, and auxiliary sensing data collected by the camera sensing module, laser sensing module, and auxiliary sensing module; the contact sensing data includes contact sensing data collected by the contact sensing module.

[0010] As a better implementation option, preferably, the fusion module described in this scheme fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data respectively, takes advantage of the strengths of the data sources and compensates for their weaknesses, and outputs obstacle information after multiple levels of redundancy. The obstacle information includes at least one of the position, speed, shape and size, category, and direction of the obstacle. The obstacle information is sent to the decision execution module; after receiving the obstacle information, the decision execution module performs corresponding defensive actions according to preset requirements.

[0011] As a preferred implementation option, preferably, the defensive action in this solution includes planning an obstacle avoidance path for the unmanned garbage sweeper to avoid collision between the garbage sweeper and obstacles; The fusion module fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data, and when outputting obstacle information, identifies the obstacle based on one of the camera perception data, laser perception data, and auxiliary perception data. When the obstacle is determined to be garbage, it is cleaned by the unmanned garbage cleaning vehicle. When cleaning cannot be completed, the contact pressure data between the obstacle and the unmanned garbage cleaning vehicle is determined based on the contact perception data collected by the contact perception module deployed thereon, and a judgment is made.

[0012] If the pressure data is less than or equal to the processable threshold of the unmanned garbage sweeper, it is determined to be a processable obstacle, and the processable obstacle is pushed to the designated area by the obstacle moving component pre-deployed on the unmanned garbage sweeper; If the pressure data exceeds the processable threshold of the unmanned garbage sweeper, it is determined to be an unprocessable obstacle, and an obstacle avoidance path for the unmanned garbage sweeper is planned to prevent the garbage sweeper from colliding with the obstacle.

[0013] As a better implementation option, preferably, the obstacle moving assembly described in this scheme is composed of a push rod, a push plate, a swing arm and a swing arm connecting rod deployed at the front end of the unmanned garbage sweeper. The two ends of the push plate are arc-shaped to prevent the obstacles from detaching during the pushing process. One end of the swing arm is fixedly connected to the push plate, and the other end of the swing arm is connected to the trolley. The midpoints of the two swing arms are connected by a swing arm connecting rod, and the midpoint of the swing arm connecting rod is connected to the movable end of the push rod. One end of the push rod is installed in the trolley. Driven by the push rod, the push plate can make circular motion around the connection point between the swing arm and the trolley. The push plate does not contact the contact plane of the pressure sensor in its initial position. Driven by the push rod, it can move to the front of the contact plane of the pressure sensor, and the push plate plane is parallel to the contact plane of the pressure sensor.

[0014] If it is determined that there is a handleable obstacle at the front of the sweeper, the push plate moves to the front of the pressure sensor contact plane. The garbage sweeper plans the path and pushes the handleable obstacle to the designated area in the server. The push plate returns to the initial position driven by the push rod and sends the movable obstacle location information to the management personnel.

[0015] As a preferred implementation option, the obstacle moving component is a part of the garbage sweeper body, and mainly relies on the power of the garbage sweeper body to push the movable obstacle to the designated area.

[0016] As a preferred implementation option, a designated area for storing movable obstacles exists within the service area. The designated area must not affect the movement of vehicles and pedestrians within the service area and must be accessible to garbage trucks.

[0017] As a better implementation option, preferably, the obstacle moving component pre-deployed on the unmanned garbage sweeper described in this solution pushes the processable obstacles to the designated area, and also feeds back the location information of the obstacles to the background management platform for the management personnel to review.

[0018] Based on the above, the present invention also provides a method for operating an unmanned garbage sweeper in a highway service area, characterized in that it applies the above-mentioned active defense system for unmanned garbage sweepers based on multi-sensor fusion.

[0019] By adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art: the present scheme collects the camera perception data in the service area in real time through the road-side monitoring module, and sends the object movement information in the service area to the fusion module; the on-board non-contact perception module collects the laser perception data, camera perception data, and auxiliary perception data of the sweeper in real time, and sends these data to the fusion module; the on-board contact perception module collects the contact perception data between the sweeper and the target garbage in real time, and sends the contact perception data to the fusion module; the fusion module fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data, outputs obstacle target information, and sends it to the decision execution module; after receiving the obstacle target information, the decision execution module makes a decision and performs corresponding defense actions to avoid the garbage sweeper from colliding with obstacles. The present scheme improves the safety of unmanned garbage sweepers working in high-speed service areas through the active defense system, device, and equipment, and avoids collisions between unmanned garbage sweepers and obstacles. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A flowchart of the implementation scheme of the active defense system module of the unmanned garbage sweeper in the high-speed service area based on multi-sensor fusion provided by the implementation example of the present invention; Figure 2 It is a schematic diagram of the communication connection relationship between the module of the active defense system of the unmanned garbage sweeper in the high-speed service area based on multi-sensor fusion and the module deployed on the unmanned garbage sweeper provided by the implementation example of the present invention; Figure 3 A data fusion structure diagram of an active defense system for unmanned garbage sweepers in high-speed service areas based on multi-sensor fusion provided by an implementation example of the present invention; Figure 4 An active defense device for unmanned garbage sweepers in high-speed service areas based on multi-sensor fusion provided by an implementation example of the present invention; Figure 5 A location diagram of an active defense device of an unmanned garbage sweeper in a high-speed service area based on multi-sensor fusion provided by an implementation example of the present invention; Figure 6 One of the active defense devices for unmanned garbage sweepers in high-speed service areas based on multi-sensor fusion provided by the implementation examples of the present invention; Figure 7 A schematic diagram of the second right front of an active defense device for an unmanned garbage sweeper in a high-speed service area based on multi-sensor fusion provided by an implementation example of the present invention; Figure 8 The left rear schematic diagram of the second active defense device of an unmanned garbage sweeper in a highway service area based on multi-sensor fusion provided as an implementation example of the present invention. DETAILED DESCRIPTION

[0022] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0023] Combination Figure 1 , Figure 2As shown, the present embodiment provides an active defense system for an unmanned garbage sweeper based on multi-sensor fusion, which is applied to an unmanned garbage sweeper in a highway service area. The unmanned garbage sweeper is equipped with a camera sensing module, a laser sensing module, an auxiliary sensing module and a contact sensing module for road surface monitoring. The highway service area is equipped with a camera for monitoring public areas, which includes: a road-side monitoring module, an on-board non-contact sensing module, an on-board contact sensing module, a fusion module and a decision execution module. S101, the roadside monitoring module is used to collect perception data captured by cameras deployed in the high-speed service area in real time, and send it to the fusion module; S102, the vehicle-mounted non-contact sensing module and contact sensing module are used to collect in real time the sensing data obtained by the camera sensing module, laser sensing module, and auxiliary sensing module deployed on the unmanned garbage sweeper, and the sensing data obtained when the unmanned garbage sweeper contacts the target garbage, and send the non-contact sensing data and contact sensing data to the fusion module; S103, the fusion module fuses the service area camera perception data, non-contact perception data, and contact perception data collected by the roadside monitoring module, the non-contact perception module, and the contact perception module according to preset requirements to output obstacle target information, and then sends it to the decision execution module; S104, after receiving the obstacle target information, the decision execution module makes a decision according to the preset requirements, and then executes a corresponding defensive action to prevent the garbage sweeper from colliding with the obstacle.

[0024] This implementation scheme can be applied to the field of unmanned garbage sweepers in highway service areas. Due to the complex environment in the service area, the vehicles in the ramp area of ​​the service area generally travel at a high speed, which is prone to collision accidents; secondly, there are many large vehicles such as trucks and hazardous chemical trucks in the service area, and the sight distance is limited, so unmanned garbage sweepers are prone to collisions with surrounding vehicles and obstacles; therefore, this scheme uses multiple sensors to perceive the situation in specific areas of the service area, and then combines the feedback from the sensors to adaptively select a response plan.

[0025] Focus on combination Figure 2 , Figure 3 As shown, as a possible implementation mode, further, the road-side monitoring module described in this scheme is communicatively connected with the security monitoring equipment deployed in the service area and the traffic monitoring equipment on the road side, both of which have cameras and are used to perform real-time monitoring of designated points in the highway service area and upload them to the local device that is communicatively connected thereto, and then use the preset algorithm to perform real-time recognition of the perception data of the scene pictures of the designated points captured by the security monitoring equipment and the traffic monitoring equipment, and monitor the movement of objects in the designated points in the service area, and then send them to the fusion module deployed in the unmanned garbage sweeper.

[0026] As a possible implementation mode, further, the vehicle-mounted non-contact sensing module described in this scheme is communicatively connected with multiple vehicle-mounted non-contact sensors deployed on the unmanned garbage cleaning vehicle, including laser radar, camera, millimeter-wave radar, ultrasonic radar, and single-line laser radar, wherein the laser radar is one of the components of the laser sensing module, the camera is one of the components of the camera sensing module, and the millimeter-wave radar, ultrasonic radar, and single-line laser radar are all components of the auxiliary sensing module.

[0027] As a possible implementation mode, further, the vehicle-mounted contact sensing module described in this solution is communicatively connected to a vehicle-mounted pressure sensor deployed on the unmanned garbage cleaning vehicle, and the vehicle-mounted pressure sensor is one of the components of the contact sensing module.

[0028] As a possible implementation mode, further, the fusion module of the present solution fuses the non-contact sensing data and contact sensing data collected by the camera sensing module, the laser sensing module, the auxiliary sensing module and the contact sensing module according to preset requirements, and the data sources complement each other, and after multi-level redundancy, finally output obstacle information based on multi-sensor fusion, wherein the obstacle information includes at least one of the position, speed, shape, size, category and orientation of the obstacle, and the obstacle information is sent to the decision execution module; Among them, the non-contact sensing data includes camera sensing data, laser sensing data, and auxiliary sensing data collected by the camera sensing module, laser sensing module, and auxiliary sensing module; the contact sensing data includes contact sensing data collected by the contact sensing module.

[0029] As a better implementation option, preferably, the fusion module described in this scheme fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data respectively, takes advantage of the strengths of the data sources and compensates for their weaknesses, and outputs obstacle information after multiple levels of redundancy. The obstacle information includes at least one of the position, speed, shape and size, category, and direction of the obstacle. The obstacle information is sent to the decision execution module; after receiving the obstacle information, the decision execution module performs corresponding defensive actions according to preset requirements.

[0030] As a preferred implementation option, preferably, the defensive action in this solution includes planning an obstacle avoidance path for the unmanned garbage sweeper to avoid collision between the garbage sweeper and obstacles; The fusion module fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data, and when outputting obstacle information, identifies the obstacle based on one of the camera perception data, laser perception data, and auxiliary perception data. When the obstacle is determined to be garbage, it is cleaned by the unmanned garbage cleaning vehicle. When cleaning cannot be completed, the contact pressure data between the obstacle and the unmanned garbage cleaning vehicle is determined based on the contact perception data collected by the contact perception module deployed thereon, and a judgment is made.

[0031] If the pressure data is less than or equal to the processable threshold of the unmanned garbage sweeper, it is determined to be a processable obstacle, and the processable obstacle is pushed to the designated area by the obstacle moving component pre-deployed on the unmanned garbage sweeper; If the pressure data exceeds the processable threshold of the unmanned garbage sweeper, it is determined to be an unprocessable obstacle, and an obstacle avoidance path for the unmanned garbage sweeper is planned to prevent the garbage sweeper from colliding with the obstacle.

[0032] Combination Figures 4 to 8 As shown, the key reference Figure 5 As shown, as a preferred implementation option, preferably, the obstacle moving assembly described in this scheme is composed of a push rod 91, a push plate 92, a swing arm 93 and a swing arm connecting rod 94 deployed at the front end of the unmanned garbage sweeper. The two ends of the push plate 92 are arc-shaped to prevent the obstacle from detaching during the pushing process. One end of the swing arm 93 is fixedly connected to the push plate 92, and the other end of the swing arm 93 is connected to the trolley. The midpoints of the two swing arms are connected by a swing arm connecting rod 94, and the midpoint of the swing arm connecting rod 94 is connected to the movable end of the push rod 91. One end of the push rod 91 is installed in the trolley. Driven by the push rod 91, the push plate 92 can make a circular motion around the connection point between the swing arm 93 and the trolley. The push plate 92 does not contact the contact plane of the pressure sensor 7 in its initial position. Driven by the push rod 91, it can move to the front of the contact plane of the pressure sensor 7, and the plane of the push plate 92 is parallel to the contact plane of the pressure sensor 7.

[0033] As a preferred implementation option, if it is determined that there is a treatable obstacle at the front end of the sweeper, the push plate 92 moves to the front of the contact plane of the pressure sensor 7, and the garbage sweeper plans the path. After the treatable obstacle is pushed to the designated area in the server, the push plate 92 returns to the initial position under the drive of the push rod 91, and the position information of the movable obstacle is sent to the management personnel.

[0034] As a preferred implementation option, the obstacle moving component is a part of the garbage sweeper body, and mainly relies on the power of the garbage sweeper body to push the movable obstacle to the designated area.

[0035] As a preferred implementation option, a designated area for storing movable obstacles exists within the service area. The designated area must not affect the movement of vehicles and pedestrians within the service area and must be accessible to garbage trucks.

[0036] As a better implementation option, preferably, the obstacle moving component pre-deployed on the unmanned garbage sweeper described in this solution pushes the processable obstacles to the designated area, and also feeds back the location information of the obstacles to the background management platform for the management personnel to review.

[0037] Figure 6 It is the road-side monitoring equipment of the service area, including the service area security monitoring equipment and roadside traffic monitoring equipment. Specifically, the equipment is the original equipment of the service area and does not need to be installed later. These devices can cover all passable areas of the unmanned garbage sweeper in the service area.

[0038] Combination Figure 7 , Figure 8 As shown, it shows the unmanned garbage sweeper on which the present solution is based. The unmanned garbage sweeper can be based on an existing sweeper and then equipped with sensors, that is, the vehicle-mounted non-contact sensors and vehicle-mounted contact sensors are installed on the vehicle body, including solid-state laser radar, wide-angle camera, telephoto camera, millimeter-wave radar, single-line laser radar, and ultrasonic radar.

[0039] in, Figures 4 to 8 In the figure, the components corresponding to the numbers are: 1. Solid-state laser radar; 2. Wide-angle camera; 3. Telephoto camera; 4. Millimeter-wave radar; 5. Single-line laser radar; 6. Ultrasonic radar; 7. Pressure sensor; 8. Road-side monitoring camera; 91. Push rod; 92. Push plate; 93. Swing arm; 94. Swing arm connecting rod.

[0040] Based on the above, the present invention also provides a method for operating an unmanned garbage sweeper in a highway service area, characterized in that it applies the above-mentioned active defense system for unmanned garbage sweepers based on multi-sensor fusion.

[0041] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An active defense system for unmanned garbage sweepers based on multi-sensor fusion, applied to unmanned garbage sweepers in highway service areas, wherein the unmanned garbage sweepers are equipped with a camera sensing module, a laser sensing module, an auxiliary sensing module and a contact sensing module for road surface monitoring, and the highway service areas are equipped with cameras for monitoring public areas, characterized in that: It includes: road-side monitoring module, vehicle-mounted non-contact sensing module, vehicle-mounted contact sensing module, fusion module, and decision-making execution module; The roadside monitoring module is used to collect perception data captured by cameras deployed in the high-speed service area in real time and send it to the fusion module; The on-board non-contact sensing module is used to collect the sensing data obtained by the camera sensing module, laser sensing module, and auxiliary sensing module deployed on the unmanned garbage sweeper in real time, and send the non-contact sensing data to the fusion module; The vehicle-mounted contact sensing module is used to collect the sensing data obtained when the unmanned garbage sweeper contacts the target garbage in real time, and send the contact sensing data to the fusion module; The fusion module fuses the service area camera perception data, non-contact perception data, and contact perception data collected by the roadside monitoring module, the non-contact perception module, and the contact perception module according to preset requirements to output obstacle target information, and then sends it to the decision execution module; After receiving the obstacle target information, the decision execution module makes a decision according to the preset requirements, and then executes the corresponding defense action to avoid the garbage sweeper from colliding with the obstacle.

2. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 1, characterized in that: The road-end monitoring module is communicatively connected to the security monitoring equipment deployed in the service area and the traffic monitoring equipment on the roadside. Both of them have cameras and are used to monitor the designated points in the highway service area in real time and upload them to the local device that is communicatively connected to them. The preset algorithm is then used to perform real-time recognition of the perception data of the designated point scene captured by the security monitoring equipment and the traffic monitoring equipment, and the movement of objects in the designated points in the service area are monitored, and then sent to the fusion module deployed in the unmanned garbage sweeper.

3. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 1, characterized in that: The vehicle-mounted non-contact sensing module is communicatively connected to multiple vehicle-mounted non-contact sensors deployed on the unmanned garbage cleaning vehicle, including laser radar, camera, millimeter-wave radar, ultrasonic radar, and single-line laser radar, wherein the laser radar is one of the components of the laser sensing module, the camera is one of the components of the camera sensing module, and the millimeter-wave radar, ultrasonic radar, and single-line laser radar are all components of the auxiliary sensing module.

4. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 1, characterized in that: The vehicle-mounted contact sensing module is communicatively connected to a vehicle-mounted pressure sensor deployed on the unmanned garbage cleaning vehicle, and the vehicle-mounted pressure sensor is one of the components of the contact sensing module.

5. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 1, characterized in that: The fusion module fuses the non-contact sensing data and contact sensing data collected by the camera sensing module, the laser sensing module, the auxiliary sensing module and the contact sensing module according to preset requirements, and the data sources complement each other and after multi-level redundancy, finally outputs obstacle information based on multi-sensor fusion, wherein the obstacle information includes at least one of the position, speed, shape, size, category and direction of the obstacle, and the obstacle information is sent to the decision execution module; Among them, the non-contact sensing data includes camera sensing data, laser sensing data, and auxiliary sensing data collected by the camera sensing module, laser sensing module, and auxiliary sensing module; the contact sensing data includes contact sensing data collected by the contact sensing module.

6. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 5, characterized in that: The fusion module fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data, respectively, and outputs obstacle information after multiple levels of redundancy by taking advantage of the strengths of the data sources and making up for their weaknesses. The obstacle information includes at least one of the position, speed, shape, size, category, and orientation of the obstacle, and the obstacle information is sent to the decision execution module; After receiving the obstacle information, the decision execution module performs corresponding defense actions according to preset requirements.

7. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 5 or 6, characterized in that: The defensive action includes planning an obstacle avoidance path for the unmanned garbage sweeper to avoid collision between the garbage sweeper and obstacles; The fusion module fuses the camera perception data, laser perception data, auxiliary perception data, and contact perception data, and when outputting obstacle information, identifies the obstacle according to one of the camera perception data, laser perception data, and auxiliary perception data. When the obstacle is determined to be garbage, it is cleaned by the unmanned garbage cleaning vehicle. When the cleaning cannot be completed, the contact pressure data between the obstacle and the unmanned garbage cleaning vehicle is determined and judged according to the contact perception data collected by the contact perception module deployed thereon; If the pressure data is less than or equal to the processable threshold of the unmanned garbage sweeper, it is determined to be a processable obstacle, and the processable obstacle is pushed to the designated area by the obstacle moving component pre-deployed on the unmanned garbage sweeper; If the pressure data exceeds the processable threshold of the unmanned garbage sweeper, it is determined to be an unprocessable obstacle, and an obstacle avoidance path for the unmanned garbage sweeper is planned to prevent the garbage sweeper from colliding with the obstacle.

8. The active defense system for unmanned garbage sweeper based on multi-sensor fusion as claimed in claim 7, characterized in that: The obstacle moving component is composed of a push rod, a push plate, a swing arm and a swing arm connecting rod deployed at the front end of the unmanned garbage sweeper. The two ends of the push plate are arc-shaped to prevent the obstacles from detaching during the pushing process. One end of the swing arm is fixedly connected to the push plate, and the other end of the swing arm is connected to the trolley. The midpoints of the two swing arms are connected by a swing arm connecting rod, and the midpoint of the swing arm connecting rod is connected to the movable end of the push rod. One end of the push rod is installed in the trolley. Driven by the push rod, the push plate can make a circular motion around the connection point between the swing arm and the trolley. The push plate does not contact the contact plane of the pressure sensor in its initial position. Driven by the push rod, it can move to the front of the contact plane of the pressure sensor, and the push plate plane is parallel to the contact plane of the pressure sensor. If it is determined that there is a treatable obstacle in front of the sweeper, the push plate moves to the front of the pressure sensor contact plane, and the garbage sweeper plans a path to push the treatable obstacle to the designated area in the server. Then, the push plate returns to the initial position driven by the push rod; The obstacle moving component is a part of the garbage sweeper body, which mainly relies on the power of the garbage sweeper body to move the movable obstacle to the designated area; The designated area for storing movable obstacles exists within the service area. The designated area must not affect the movement of vehicles and pedestrians in the service area and must be accessible by garbage trucks.

9. The active defense system for unmanned garbage cleaning vehicles based on multi-sensor fusion as claimed in claim 7, characterized in that: The obstacle moving component pre-deployed on the unmanned garbage sweeper pushes the processable obstacles to the designated area and also feeds back the location information of the obstacles to the background management platform for the management personnel to review.

10. A method for operating an unmanned garbage sweeper in a highway service area, characterized in that: The invention is applied to the active defense system for unmanned garbage sweeper based on multi-sensor fusion as described in any one of claims 1 to 9.

Images