Planetary probe vehicle

Abstract

The invention discloses a planetary probe vehicle, and belongs to the technical field of spaceflight. The planetary probe vehicle comprises a moving system, an intelligent sensing system, a control system and a power system, the moving system is used for moving the planetary probe vehicle and mounting and bearing other components of the planetary probe vehicle, and the intelligent sensing system is used for collecting visual information and depth information in front of the planetary probe vehicle; the control system is used for receiving the information data collected by the intelligent sensing system and sending a mobile control instruction to the mobile system after data processing, and the power system is used for supplying power to the mobile system and the control system. Wherein the control system comprises an embedded computer, a single chip microcomputer and an electronic speed regulator, an autonomous navigation module is arranged in the embedded computer, a navigation path can be dynamically generated by utilizing environmental data collected by the intelligent sensing system in real time, and the problem that an existing planetary probe vehicle depends on a multi-source sensor and a navigation pre-stored map is solved.

Description

Technical Field

[0001] The present invention relates to the field of space technology, and particularly to a planetary rover. Background Art

[0002] Extraterrestrial celestial body exploration, as an important practice for humans to explore the universe, has established a technical system covering multiple modes such as flyby, orbiting, landing and roving through more than 200 exploration missions. Among them, landing and roving, with its ability to directly obtain the material composition, geological structure and environmental parameters of the celestial body surface, has become an irreplaceable technical means in deep space exploration missions, laying the foundation for subsequent resource development and manned landing. As the core platform of the landing and roving mission, the planetary rover can directly obtain in-situ data on the celestial body surface through autonomous roving and provide key technical support for sampling return and manned landing missions. The navigation algorithm of the planetary rover is the core technical support for the planetary rover to achieve autonomous survival and scientific exploration in an unknown extraterrestrial environment, providing a solid guarantee for the successful completion of the exploration mission of the planetary rover.

[0003] However, there are currently two common problems in the research and development of planetary rovers: First, the strict mass limitation of space launch requires the rover to achieve light weight, high integration, high passability and stability. For the sensor system that supports the navigation of the planetary rover, on the one hand, the vehicle body needs to provide a stable and reliable working environment for it. On the other hand, it also needs to have the characteristics of high integration, light weight and reliability. The existing sensing systems of planetary rovers generally use multi-source sensors arranged dispersedly to sense the environment around the vehicle body. Its layout is complex, and it is easy to have a failure in information collection of a certain line. The system coupling effect will cause a reduction in reliability and complex data processing. Second, the existing navigation systems of planetary rovers generally rely on pre-bound high-precision digital maps, which are prone to terrain misjudgment and motion failure in uncharted areas.

[0004] Therefore, it is necessary to provide a planetary rover to solve the above problems. Summary of the Invention

[0005] Technical Problems to be Solved In order to avoid the deficiencies of the prior art, the present invention provides a planetary rover. Through the collaborative design of the vehicle body hardware structure and the autonomous navigation algorithm for real-time environment perception, a planetary rover with the ability of autonomous navigation in an unknown environment is constructed to solve the technical defects of the existing planetary rover's dependence on multi-source sensors and navigation pre-stored maps.

[0006] The technical solution of the present invention is: A planetary rover, comprising: A mobile system, used for the movement of the planetary rover and the installation and bearing of the remaining components of the planetary rover; An intelligent sensing system, used for collecting visual information and depth information in front of the planetary rover; A control system for receiving information data collected by an intelligent perception system, performing data processing, and sending a movement control instruction to a mobile system; And a power system for supplying power to the mobile system and the control system; Among them, the control system includes an embedded computer, a single-chip microcomputer, and an electronic speed governor. An autonomous navigation module is provided in the embedded computer. The autonomous navigation module is used to calculate and process the information data from the intelligent perception system to form a navigation control instruction and send it to the single-chip microcomputer; the single-chip microcomputer is used to convert the control instruction into an electrical signal and transmit it to the electronic speed governor; the electronic speed governor is used to send a control signal to the drive unit in the mobile system.

[0007] A further technical solution of the present invention is that the mobile system includes: A lower chassis, with wheels installed at the front and rear four corners of the lower chassis. The wheels are connected to the lower chassis through a suspension damping mechanism, and an upper chassis is installed above the suspension damping mechanism; among them, the suspension damping mechanism is used for damping during the movement of the planetary rover, and the upper chassis and the lower chassis are used to install and carry the remaining components of the planetary rover; A drive unit, which includes a motor and a steering gear, both installed on the lower chassis. Among them, the motor is used to drive the wheels to rotate through a transmission component, and the steering gear is used to drive the two front wheels at the front end of the movement to turn; both the motor and the steering gear are electrically connected to the electronic speed governor.

[0008] A further technical solution of the present invention is that the transmission component includes a belt pulley component, a transmission shaft, a front gearbox, a rear gearbox, and a half shaft. The output end of the motor is connected to the middle of the transmission shaft through the belt pulley component. One end of the transmission shaft is in transmission connection with the power input end of the front gearbox installed in the middle of the front end of the lower chassis, and the other end of the transmission shaft is in transmission connection with the power input end of the rear gearbox installed in the middle of the rear end of the lower chassis. The two output ends of the front gearbox provide power for the two front wheels through the half shaft, and the two output ends of the rear gearbox provide power for the two rear wheels through the half shaft.

[0009] A further technical solution of the present invention is that a wheel steering mechanism is installed on the two front wheels, the output end of the steering gear is connected to the input end of the wheel steering mechanism, and the wheel steering mechanism is used for the steering control of the two front wheels.

[0010] A further technical solution of the present invention is that the intelligent perception system includes a depth camera, which is installed above the upper chassis through a connecting rod. The depth camera is used to collect visual information and depth information in front of the planetary rover, and the depth camera is electrically connected to the embedded computer.

[0011] A further technical solution of the present invention is that the embedded computer is installed on the upper chassis, the single-chip microcomputer and the electronic speed controller are installed on the lower chassis, the embedded computer is electrically connected to the single-chip microcomputer, and the single-chip microcomputer is electrically connected to the electronic speed controller.

[0012] A further technical solution of the present invention is that the power system includes a first lithium battery and a second lithium battery. The first lithium battery is installed on the upper chassis and is used to supply power to the embedded computer. A transformer is connected in series in the power supply circuit between the first lithium battery and the embedded computer. The second lithium battery is installed on the lower chassis. The second lithium battery is used to supply power to the single-chip microcomputer, the electronic speed controller, the servo and the motor. The second lithium battery is electrically connected to the power switch, and the power supply of the second lithium battery is controlled by the power switch.

[0013] A further technical solution of the present invention is that a vehicle body protection cover is installed on the upper chassis for protecting the power system and the control system. Slots are symmetrically arranged at the bottom of the side plates on both sides of the vehicle body protection cover for plugging with both sides of the upper chassis. The rear partition of the vehicle body protection cover is a detachable structure and is inserted and connected to the main body part of the vehicle body protection cover for opening the vehicle body protection cover. Among them, the connecting rod passes through the top of the vehicle body protection cover.

[0014] A planetary exploration vehicle autonomous navigation algorithm includes: Obtaining the current state data of the planetary exploration vehicle; the current state data includes: the distance between the planetary exploration vehicle and the end target; the distance between the planetary exploration vehicle and the nearest obstacle; the angle between the forward direction of the planetary exploration vehicle and the line connecting the planetary exploration vehicle to the target; the angle between the forward direction of the planetary exploration vehicle and the line connecting the planetary exploration vehicle to the nearest obstacle. Setting a reward function, pre-training a network model based on the SAC reinforcement learning algorithm, inputting the current state data into the pre-trained network model to obtain the forward direction change amount of the planetary exploration vehicle, and controlling the traveling direction of the planetary exploration vehicle according to the forward direction change amount until the planetary exploration vehicle reaches the end point.

[0015] A further technical solution of the present invention is that the set reward function is:

[0016] In the formula, is the target reward, which is used to reward the behavior of the planetary exploration vehicle reaching the end point and is defined as: ; In the formula, is the distance threshold for determining that the planetary exploration vehicle reaches the end point; is the penalty reward, which is used to punish the behavior of the planetary exploration vehicle colliding with an obstacle or exceeding the exploration boundary and is defined as: ; In the formula, is beyond the boundary; is the gradient reward, which is set based on the distance gradient of the planetary rover to the target and is used to guide the planetary rover to explore in the direction of the target. It is defined as:

[0017] is the direction reward, which is used to make the forward direction of the planetary rover more inclined to approach the target. The expression is: ; is the step reward, , which is used to reduce the total number of iterations to obtain a shorter path.

[0018] The beneficial effects of the present invention are as follows: The present invention innovatively proposes a modular planetary rover system. Through the overall mechanism design of the planetary rover, the depth camera is fixedly installed at the top of the connecting rod, and the connecting rod is installed about 60 cm higher than the upper chassis, which can ensure its field of vision. The depth camera provides the embedded computer with the visual information and depth information in front of the driving rover, providing data for the calculation of the traveling trajectory and avoiding the use of a multi-source sensing system.

[0019] The embedded computer in the present invention integrates an autonomous navigation algorithm based on reinforcement learning, which has the functions of integrated autonomous navigation and guidance. By installing a depth camera on the planetary rover, the depth camera can collect the visual information and depth information in front of the planetary rover, and through the autonomous navigation algorithm integrated in the embedded computer, it can dynamically generate a navigation path using the environmental data collected by the depth camera in real time, successfully breaking through the bottleneck of the traditional method's dependence on the pre-stored map. This technical system significantly improves the ground verification efficiency and provides an economical and reliable engineering solution for deep space exploration missions.

[0020] Compared with the traditional planetary rover solution, the autonomous navigation method developed for the planetary rover designed in the present invention realizes an integrated autonomous navigation and guidance solution for the planetary rover that only depends on the camera, avoids the dependence on complex and redundant sensors, and improves the reliability of the system.

[0021] The wheel of the present invention adopts an all-terrain rubber tire and is connected to the vehicle body chassis through a suspension damping structure. The suspension damping structure adopts a damping mechanism, and realizes vibration attenuation through mechanical energy absorption design, ensuring the operation stability of on-vehicle equipment, enabling the planetary exploration vehicle of the present invention to adapt to complex road conditions and having good passability. Compared with the traditional planetary vehicles with caterpillar structures and mechanical leg structures, the vehicle body weight is greatly reduced, while stability and passability are taken into account at the same time.

[0022] The vehicle body chassis is formed by a double-layer structure of an upper chassis and a lower chassis, realizing the layout of the moving system, intelligent perception system, control system and power system of the planetary exploration vehicle, increasing the bearing and installation area, and at the same time reducing the structural size of the planetary exploration vehicle. At the same time, a hollowing design is carried out for the areas with less chassis stress and non-installation bearing areas, which is more conducive to reducing the vehicle body weight.

[0023] The vehicle body protective cover and the upper chassis adopt an insertion structure design, which is convenient for installation. At the same time, the rear partition of the vehicle body protective cover and the main body of the vehicle body protective cover adopt an insertion structure design, which is convenient to open the vehicle body protective cover for internal inspection and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0025] Figure 1 It is a schematic diagram of the overall effect of the planetary exploration vehicle of the present invention; Figure 2 It is an axonometric schematic diagram of the layout structure of the lower chassis of the planetary exploration vehicle of the present invention; Figure 3 It is a top view schematic diagram of the layout structure of the lower chassis of the planetary exploration vehicle of the present invention; Figure 4 It is a schematic diagram of the layout structure of the upper chassis of the planetary exploration vehicle of the present invention; Figure 5 It is a schematic diagram of the structure of the vehicle body protective cover in the present invention; Figure 6 It is a schematic diagram of the wiring connection situation of the lower chassis in the present invention; Figure 7 It is a schematic diagram of the wiring connection situation of the upper chassis in the present invention; Figure 8 It is a picture of the coordinate system construction in the training environment of the planetary exploration vehicle of the present invention; Figure 9 It is a schematic diagram of the action space during the test experiment process of the planetary exploration vehicle of the present invention; Figure 10 This is the moving path during the test experiment of the planetary rover of the present invention; Figure 11 This is the action value adopted by the planetary rover during the test experiment of the planetary rover of the present invention.

[0026] In the figure: 1. Wheel, 2. Second lithium battery, 3. Electronic speed controller, 4. Motor, 5. Steering gear, 6. Power switch, 7. Suspension shock absorption mechanism, 8. Anti-collision protection structure, 9. Lower chassis, 10. Depth camera, 11. Connecting rod, 12. Upper chassis, 13. Embedded computer, 14. First lithium battery, 15. Transformer, 16. Vehicle body protection cover, 17. Through hole, 18. Rear partition slot, 19. Rear partition of the protection shell, 20. Slot, 21. First connection hole, 22. Single-chip microcomputer, 23. Belt pulley assembly, 24. Transmission shaft, 25. Front gearbox, 26. Rear gearbox, 27. Wheel steering mechanism, 28. Second connection hole. Specific embodiments

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] An embodiment of the planetary rover of the present invention is a simulation prototype for ground test verification of the planetary rover. Due to the high cost of spacecraft launch, the ground verification link must bear higher reliability verification requirements. Therefore, various ground test verifications need to be carried out first through the simulation prototype. After passing the verification, the officially used planetary rover will be produced according to the prototype.

[0029] The present invention mainly aims at the technical defects of the existing planetary rover relying on multi-source sensors and navigation pre-stored maps. Through the overall vehicle body structure design and the design of the autonomous navigation and guidance algorithm for real-time environment perception, a simulation prototype of the planetary rover with the integrated function of autonomous navigation and guidance is developed. The autonomous navigation algorithm developed by the present invention uses the environmental data collected in real time by the depth camera to dynamically generate the navigation path, successfully breaking through the dependence bottleneck of the traditional method on the pre-stored map and multi-sensors. At the same time, by reducing the number of sensors in the rover system, on the one hand, the problem of reduced reliability caused by the system coupling effect is effectively solved, and on the other hand, the load mass and volume of the planetary rover are reduced, providing conditions for the planetary rover to carry other effective payloads.

[0030] The planetary exploration vehicle includes a mobile system, an intelligent perception system, a control system, and a power system. Among them, the mobile system is mainly used for the movement of the planetary exploration vehicle and the installation and bearing of the other components of the planetary exploration vehicle; the intelligent perception system is used to collect visual information and depth information in front of the planetary exploration vehicle and transmit the collected information to the control system for processing; the control system is used to receive the information data collected by the intelligent perception system, and after data processing, send a movement control instruction to the mobile system; the power system is used to supply power to the mobile system and the control system.

[0031] Specifically, referring to Figures 1-5 , the mobile system, as the main body for bearing and movement of the planetary exploration vehicle, mainly includes a vehicle body chassis, wheels 1, a suspension damping mechanism 7, a transmission component, a wheel steering mechanism 27, and a drive unit.

[0032] The vehicle body chassis is divided into two layers, including a lower chassis 9 and an upper chassis 12. Four wheels 1 are installed at the four corners of the front and rear of the lower chassis 9. The two wheels in the front of the traveling direction are the front wheels, and the two wheels in the rear of the traveling direction are the rear wheels. The two front wheels are equipped with steering components for realizing steering during traveling. The wheels 1 are connected to the lower chassis 9 through the suspension damping mechanism 7. There are two sets of suspension damping mechanisms 7, which are respectively installed at the front and rear ends of the lower chassis 9. The upper chassis 12 is installed above the two sets of suspension damping mechanisms 7. Four first connection holes 21 are provided at the top of the bracket of the suspension damping mechanism 7, and four second connection holes 28 are provided on the upper chassis 12 corresponding to the first connection holes 21. The upper chassis 12 is fixedly connected to the suspension damping mechanism 7 by installing fasteners in the corresponding connection holes. The suspension damping mechanism 7 is used for damping during the traveling of the planetary exploration vehicle, and the upper chassis 12 and the lower chassis 9 are used for installing and bearing the other components of the planetary exploration vehicle.

[0033] In this embodiment, the lower chassis 9 is integrally cut by a numerically controlled machine tool. Its length and width are 80 cm and 27 cm respectively, the ground clearance is 12 cm, the mass is about 25 kg, and the position with relatively small chassis stress adopts a hollow design to reduce the weight. The wheels 1 adopt all-terrain rubber tires with a radius of 10 cm.

[0034] The drive unit includes a motor 4 and a servo 5, both of which are installed on the lower chassis 9. The motor 4 is used to drive the wheels 1 to rotate through the transmission component, and the servo 5 is used to drive the two front wheels at the front end of the traveling to turn. Specifically, as Figure 2 , Figure 3As shown in the figure, the transmission assembly mainly includes a pulley assembly 23, a transmission shaft 24, a front gearbox 25, a rear gearbox 26 and half shafts. The output end of the motor 4 is connected to the middle of the transmission shaft 24 through the pulley assembly 23. The pulley assembly uses a belt and a pulley in the prior art and is used to transmit the rotation of the output shaft of the motor 4 to the rotation of the transmission shaft 24. The transmission shaft 24 is arranged on the lower chassis 9 along the traveling direction of the vehicle body and is located in the middle of the two side wheels 1. The front end of the transmission shaft 24 is drivingly connected to the power input end of the front gearbox 25, and the rear end of the transmission shaft 24 is drivingly connected to the power input end of the rear gearbox 26. The front gearbox 25 is installed in the middle of the front end of the lower chassis 9, and the rear gearbox 26 is installed in the middle of the rear end of the lower chassis 9. Both the front and rear gearboxes play a supporting role in the middle of the suspension damping mechanism 7 at the installation location. The two output ends on both sides of the front gearbox 25 provide power for the two front wheels through half shafts, and the two output ends on both sides of the rear gearbox 26 provide power for the two rear wheels through half shafts (the half shafts are not shown in the figure). The rotation of the half shafts at the output end of the gearbox is driven by the transmission shaft 24, and then the wheels 1 are driven to rotate. The transmission shaft 24, the gearbox and the half shafts adopt existing mature technologies and will not be elaborated here. The output end of the steering gear 5 is connected to the input end of the wheel steering mechanism 27. The wheel steering mechanism 27 is connected to the two front wheels and is used for the steering control of the two front wheels. The action of the wheel steering mechanism 27 is driven by the rotation of the output arm of the steering gear 5, and then the front wheels are driven to turn a certain angle.

[0035] The intelligent perception system includes a depth camera 10. The depth camera 10 is installed above the upper chassis 12 through a connecting rod 11. The connecting rod 11 is detachably installed perpendicular to the upper chassis 12, and the depth camera 10 is installed at the top of the connecting rod 11 far from the upper chassis 12. The depth camera 10 is used to collect visual information and depth information in front of the planetary rover. In order to ensure a good vision effect, it is about 60 cm higher than the upper chassis 12. The data output end of the depth camera 10 is electrically connected to the control system and is used to transmit the collected data to the control system.

[0036] The control system includes an embedded computer 13, a single-chip microcomputer 22, and an electronic speed governor 3. The embedded computer 13 is installed on the upper chassis 12, and the single-chip microcomputer 22 and the electronic speed governor 3 are installed on the lower chassis 9. The embedded computer 13 is electrically connected to the depth camera 10 and is also electrically connected to the single-chip microcomputer 22, and the single-chip microcomputer 22 is electrically connected to the electronic speed governor 3. The embedded computer 13 is the central hub of the control system. The embedded computer 13 is equipped with an autonomous navigation module. The embedded computer 13 receives information data from the intelligent perception system, converts it into a movement control instruction for the planetary rover through its autonomous navigation module, and sends it to the single-chip microcomputer 22 in a specific encoded format. The single-chip microcomputer 22 converts the received control instruction into an electrical signal and transmits it to the electronic speed governor 3. The electronic speed governor 3 is electrically connected to the motor 4 and is also electrically connected to the steering gear 5. The electronic speed governor 3 is used to control the operation of the motor 4 and the steering gear 5, including speed control and start / stop control.

[0037] The power system includes a first lithium battery 14 and a second lithium battery 2. The first lithium battery 14 is installed on the upper chassis 12 and is used to supply power to the embedded computer 13. A transformer 15 is connected in series in the power supply circuit between the first lithium battery 14 and the embedded computer 13 for voltage stabilization. The second lithium battery 2 is installed on the lower chassis 9. The second lithium battery 2 is used to supply power to the single-chip microcomputer 22, the electronic speed governor 3, the steering gear 5, and the motor 4. The second lithium battery 2 is electrically connected to the power switch 6, and the power switch 6 controls the on / off of the power supply of the second lithium battery 2. Both lithium batteries are fixed to the corresponding chassis with glue. As Figure 6 、 Figure 7 shown, Figure 6 in which, C2-3 represents the connection between the second lithium battery 2 and the electronic speed governor 3; C2-4 represents the connection between the second lithium battery 2 and the motor 4; C2-22 represents the connection between the second lithium battery 2 and the single-chip microcomputer 22; C2-5 represents the connection between the second lithium battery 2 and the steering gear 5; C3-4 represents the connection between the electronic speed governor 3 and the motor 4; C3-5 represents the connection between the electronic speed governor 3 and the steering gear 5; C6-2 represents the connection between the power switch 6 and the second lithium battery 2; C13-22 represents the connection between the embedded computer 13 and the single-chip microcomputer 22, which is connected between the two layers of the lower chassis 9 and the upper chassis 12, and the C13-22 connection uses a MicoreUSB data cable; C22-3 represents the connection between the single-chip microcomputer 22 and the electronic speed governor 3. Figure 7 in which, C14-15 represents the connection between the first lithium battery 14 and the transformer 15; C15-13 represents the connection between the transformer 15 and the embedded computer 13; C10-13 represents the connection between the depth camera 10 and the embedded computer 13, and the C10-13 connection uses a type-C data cable for connection. Except for C13-22 and C10-13, the rest of the connecting wires in this embodiment are not required, and DuPont wires are recommended. Figure 6, Figure 7 The arrow direction in it indicates the transmission direction of signals or energy.

[0038] To protect the power system and control system, a vehicle body protection cover 16 is installed on the upper chassis 12. The vehicle body protection cover 16 is made of photosensitive resin material through 3D printing. The vehicle body protection cover 16 consists of a main body part and a rear partition. Its main body part is a shell structure with an open bottom and rear side. On the inner sides of the bottoms of the two side plates of the main body part, slots 20 are symmetrically arranged for plugging with the two sides of the upper chassis 12; a through hole 17 for passing through the connecting rod 11 is arranged on the top plate of the main body part. At the open rear end of the main body part, a rear partition slot 18 is vertically arranged perpendicular to the upper chassis 12, and the rear partition 19 is vertically inserted into the rear partition slot 18 to close the open rear side of the main body part. The plug-in connection method of the rear partition 19 is convenient for disassembly and assembly, facilitating the operation of the internal control system and power system.

[0039] During the working process of the planetary rover, first, the depth camera 10 of the intelligent perception system obtains environmental information to get the target position information and obstacle information in the environment. The depth camera 10 transmits the information to the embedded computer 13; the autonomous navigation module of the embedded computer 13 runs its integrated autonomous navigation algorithm. After calculation, the movement control instruction of the planetary rover is obtained and transmitted to the lower single-chip microcomputer 22 through serial communication. The single-chip microcomputer 22 converts the control instruction into an electric pulse signal and transmits it to the electronic speed controller 3; then, the electronic speed controller 3 finally controls the motor 4 and the steering gear 5 to realize the movement of the planetary rover. The motor 4 drives the wheels 1 to rotate, and the steering gear 5 controls the steering of the front wheels. Finally, the entire system completes the control of the planetary rover through this information flow.

[0040] An autonomous navigation algorithm for a planetary rover provided by this embodiment is integrated in the autonomous navigation module of the embedded computer 13. This algorithm obtains the current state data of the planetary rover, inputs the current state data into a pre-trained network model based on the SAC reinforcement learning algorithm, obtains the forward direction change amount of the planetary rover, and controls the traveling direction of the planetary rover according to the forward direction change amount until the planetary rover reaches the end point.

[0041] This algorithm is an integrated autonomous navigation and guidance method for a planetary rover based on SAC reinforcement learning, which is used for the autonomous navigation of the planetary rover in the scenario of missing global information. This method is developed based on the SAC reinforcement learning algorithm. The network structure adopts a fully connected neural network, the activation function uses ReLU, the state space is selected through the intelligent perception system of the planetary rover, and the action space is selected as the forward direction change amount of the planetary rover. The reward function is designed as a reward function that guides the planetary rover to move towards the target and reduces collisions. The specific training method of the network model is as follows: As Figure 8As shown, set up a training environment, establish a training environment coordinate system, and randomly set a starting point and an ending point in the training environment coordinate system. The positions of the starting point and the ending point are not fixed and change randomly. The starting point is randomly and uniformly distributed within and . The determination range of the ending point is , and the position is randomly and uniformly distributed within and . Set three obstacles. The change range of the x coordinate of the first obstacle is , the change range of the y coordinate is , and the change range of the radius is . The change range of the x coordinate of the second obstacle is , the change range of the y coordinate is , and the change range of the radius is . The change range of the x coordinate of the third obstacle is , the change range of the y coordinate is , and the change range of the radius is .

[0042] Select the current state data parameters of the planetary rover as:

[0043] In the formula, represents the distance between the planetary rover and the end target, with the unit of m; represents the distance between the planetary rover and the nearest obstacle, with the unit of m; represents the angle between the forward direction of the planetary rover and the line connecting the planetary rover to the target, with the unit of degree; represents the angle between the forward direction of the planetary rover and the line connecting the planetary rover to the nearest obstacle, with the unit of degree; , , , are all provided by the intelligent perception system. Since the speed change range of the planetary rover is small, the speed is not selected as one of the actions but regarded as a constant. Therefore, only the change amount of the forward direction of the planetary rover is selected as the action in this algorithm, as shown in Figure 9 . Figure 9 In , is the angle between the speed direction of the planetary rover and the forward axis of the planetary rover, and

[0044] represent the speeds of the planetary rover before and after executing the action respectively.

[0045] In the formula, is the target reward, which is used to reward the behavior of the planetary rover reaching the end point. For the behavior of reaching the target point, a reward is given. The purpose is to ensure that the exploration can finally reach the end point, and it is defined as: ; In the formula, is the distance threshold for determining that the planetary rover reaches the end point; is the penalty reward, which is mainly used to punish the behavior of the planetary rover colliding with obstacles or exceeding the exploration boundary. For the behavior of entering the obstacle range or exceeding the exploration boundary, based on the penalty value, its behavior is constrained, and it is defined as: ; In the formula, is exceeding the boundary; To accelerate the training process and obtain a faster convergence speed, a reward set based on the distance gradient of the agent to the target is introduced into the reward function , and it is defined as:

[0046] is the direction reward, which is the reward given according to the advancing direction of the planetary rover. It is used to make the advancing direction of the planetary rover more inclined to approach the target, thereby reducing the path length. At the same time, it is also to enable the planetary rover to reach the end point faster. The expression is: ; is the step reward, and is set, which is used to reduce the total number of iterations to obtain a shorter path.

[0047] The SAC reinforcement learning algorithm (i.e., the maximum entropy reinforcement learning algorithm) is used for training. The network structure adopts fully connected layers, with 3 hidden layers and 128 nodes. After the above steps are set, the algorithm is trained in a virtual scenario, and the trained algorithm network is saved, that is, the network model based on the SAC reinforcement learning algorithm, and the network can be deployed to an embedded computer for use.

[0048] The following gives the specific application method of the present invention in combination with a specific application scenario: Before using the planetary rover, it is necessary to charge each lithium battery first. The standard voltage of the charger is 25.2V. After the two lithium batteries are fully charged, press the power switch 6 to start the first lithium battery 2 to supply power to the single-chip microcomputer 22, electronic speed controller 3, servo 5 and motor 4 on the lower chassis 9. At the same time, connect the power supply of the first lithium battery 14, turn on the embedded computer 13, and run the relevant program to generate control instructions. After the lithium battery of the planetary rover is fully charged, it can support the test work of the planetary rover for about 2 hours. When the power is insufficient, the moving speed of the planetary rover will drop significantly.

[0049] The conditions in the experimental scenario are set as follows: the target is located at (3.25m, 3.00m). The obstacles are selected as two isosceles right triangle cylinders with a height of 1m and a right-angled side length of 0.3m. Their central positions are (3.00m, 1.95m) and (1.30m, 2.21m) respectively. When the distance between the planetary vehicle and the target is less than 0.4m, it is considered that the target location has been successfully reached. The starting positions of the planetary rovers are randomly distributed.

[0050] In this experiment, the only sensor used is the depth camera of the planetary rover, which is used for target recognition and obstacle perception. The specific algorithms for target recognition and obstacle perception adopt existing technologies and will not be elaborated here. Using the depth camera, we can obtain the distance from the planetary rover prototype to the target, the line-of-sight angle between the target and the planetary rover prototype, and the position information of the obstacles around the planetary rover.

[0051] The main function of the path planning algorithm developed by the present invention running in the embedded computer is to receive the target information and obstacle information in the environment output by the target recognition and obstacle detection algorithms, obtain the control instructions for the planetary rover, convert them into the required moving speed and moving direction, and then send instructions to the lower computer (i.e., the single-chip microcomputer). The instructions are sent through serial communication, and the baud rate of the serial communication is 115200. The format of the instruction before encoding is "45n100", where "45" represents half of the angle between the wheel direction of the planetary rover and the line connecting the wheels, and its range of variation is (30, 60). "45" means the wheels are facing directly forward. If this value is "30", it means the wheels are deflected 30° to the right; "100" is the percentage of the current output power of the motor compared to the maximum output power, and its range of variation is (0, 100). "100" means the motor output power is the maximum. If this value is "000", it means to stop the output; "n" is a detection character used to distinguish the two values. After obtaining the specific control quantity, the embedded computer encodes the instruction and sends it to the single-chip microcomputer through serial communication.

[0052] The single-chip microcomputer receives the encoded control instructions from the embedded computer through serial communication. First, it decodes the instructions to restore them to the standard instruction form, such as "45n100". Subsequently, the single-chip microcomputer will adjust the levels of the pins corresponding to the servo and the motor respectively according to the received instructions. The pins corresponding to the servo and the motor are connected to the electronic speed controller through Dupont wires. After the levels are converted by the electronic speed controller, they are sent to the servo and the motor respectively to achieve the control of the movement direction and speed of the planetary rover simulation prototype.

[0053] This process repeats until the planetary rover simulation prototype reaches the target range. The algorithm in the embedded computer issues an instruction of "45n000". The planetary rover straightens its wheels and stops working. Subsequently, the overall algorithm ends its operation and the test is completed. The movement trajectory of the planetary rover during the test is as Figure 10 shown, which proves the effectiveness of the algorithm provided by the present invention. The action execution situation of the planetary rover during the test is as Figure 11 shown, which shows the change of the action of the planetary rover with the number of iterations. The action refers to the change of the tire angle of the planetary rover at each iteration.

[0054] After testing, the planetary rover can reliably run the algorithms that need to be verified by physical simulation. At the same time, the related equipment it is equipped with can support the simulation tests of the vast majority of mainstream planetary rover algorithms. And the model prototype of the planetary rover has a certain degree of expandability and can be modified or equipped with other devices, such as an odometer and other devices, to better meet the requirements of relevant algorithm tests.

[0055] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A planetary exploration vehicle, characterized in that, Including: A mobile system for the movement of a planetary rover and the installation and bearing of the remaining components of the planetary rover; An intelligent perception system for collecting visual information and depth information in front of the planetary rover; A control system for receiving the information data collected by the intelligent perception system, performing data processing, and sending movement control instructions to the mobile system; And a power system for supplying power to the mobile system and the control system; Among them, the control system includes an embedded computer, a single-chip microcomputer, and an electronic speed controller. An autonomous navigation module is provided in the embedded computer. The autonomous navigation module is used to calculate and process the information data from the intelligent perception system to form navigation control instructions and send them to the single-chip microcomputer; The single-chip microcomputer is used to convert the control instructions into electrical signals and transmit them to the electronic speed controller; the electronic speed controller is used to send control signals to the drive unit in the mobile system.

2. The planetary rover according to claim 1, wherein The mobile system includes: A lower chassis, with wheels installed at the front and rear four corners of the lower chassis. The wheels are connected to the lower chassis through a suspension damping mechanism, and an upper chassis is installed above the suspension damping mechanism; among them, the suspension damping mechanism is used for damping during the movement of the planetary rover, and the upper chassis and the lower chassis are used for installing and bearing the remaining components of the planetary rover; A drive unit, which includes a motor and a steering gear, both installed on the lower chassis. Among them, the motor is used to drive the wheels to rotate through a transmission component, and the steering gear is used to drive the two front wheels at the front end of the movement to turn; both the motor and the steering gear are electrically connected to the electronic speed controller.

3. The planetary rover according to claim 2, characterized in that, The transmission component includes a belt pulley component, a transmission shaft, a front gearbox, a rear gearbox, and a half shaft. The output end of the motor is connected to the middle of the transmission shaft through the belt pulley component. One end of the transmission shaft is in transmission connection with the power input end of the front gearbox installed in the middle of the front end of the lower chassis, and the other end of the transmission shaft is in transmission connection with the power input end of the rear gearbox installed in the middle of the rear end of the lower chassis. The two output ends of the front gearbox provide power for the two front wheels through the half shaft, and the two output ends of the rear gearbox provide power for the two rear wheels through the half shaft.

4. The planetary rover according to claim 2, characterized in that, Wheel steering mechanisms are installed on the two front wheels, and the output end of the steering gear is connected to the input end of the wheel steering mechanism. The wheel steering mechanism is used for the steering control of the two front wheels.

5. The planetary rover according to claim 2, characterized in that, The intelligent perception system includes a depth camera. The depth camera is installed above the upper chassis through a connecting rod. The depth camera is used to collect visual information and depth information in front of the planetary rover, and the depth camera is electrically connected to the embedded computer.

6. The planetary rover according to claim 2, characterized in that, The embedded computer is installed on the upper chassis, the single-chip microcomputer and the electronic speed controller are installed on the lower chassis, the embedded computer and the single-chip microcomputer are electrically connected, and the single-chip microcomputer and the electronic speed controller are electrically connected.

7. The planetary rover according to claim 2, characterized in that, The power system includes a first lithium battery and a second lithium battery. The first lithium battery is installed on the upper chassis and is used to supply power to the embedded computer. A transformer is connected in series in the power supply circuit between the first lithium battery and the embedded computer; the second lithium battery is installed on the lower chassis. The second lithium battery is used to supply power to the single-chip microcomputer, the electronic speed controller, the steering gear, and the motor. The second lithium battery is electrically connected to the power switch, and the power supply switch of the second lithium battery is controlled through the power switch.

8. The planetary rover according to claim 5, wherein A vehicle body protection cover is installed on the upper chassis for protecting the power system and the control system. Slots are symmetrically provided at the bottoms of the side plates on both sides of the vehicle body protection cover for plugging with both sides of the upper chassis. The rear partition of the vehicle body protection cover is a detachable structure and is inserted and connected with the main body part of the vehicle body protection cover for opening the vehicle body protection cover. Among them, the connecting rod passes through the top of the vehicle body protection cover.

9. An autonomous navigation algorithm for a planetary rover, characterized in that, Including: Obtaining the current state data of the planetary rover; The current state data includes: the distance between the planetary rover and the end target; the distance between the planetary rover and the nearest obstacle; the angle between the forward direction of the planetary rover and the line connecting the planetary rover to the target; the angle between the forward direction of the planetary rover and the line connecting the planetary rover to the nearest obstacle. Setting a reward function, pre-training a network model based on the SAC reinforcement learning algorithm, inputting the current state data into the pre-trained network model to obtain the forward direction change amount of the planetary rover, and controlling the traveling direction of the planetary rover according to the forward direction change amount until the planetary rover reaches the end point.

10. The autonomous navigation algorithm for a planetary rover according to claim 9, wherein The set reward function is: In the formula, is the target reward, which is used to reward the behavior of the planetary rover reaching the end point and is defined as: ； In the formula, is the distance threshold for determining that the planetary rover has reached the end point; For penalty rewards, which are used to penalize the behavior of a planetary rover hitting an obstacle or exceeding the exploration boundary, it is defined as: ； In the formula, is out of bounds; It is the gradient reward, which is set based on the distance gradient of the planetary rover to the target and is used to guide the planetary rover to explore in the direction of the target. It is defined as: It is a direction reward, which is used to make the forward direction of the planetary rover more inclined to approach the target. The expression is as follows: ； For step rewards , which is used to reduce the total number of iterative steps to obtain a shorter path.

Images