A swordsmanship training device

Abstract

The application relates to a sword training device and belongs to the technical field of training devices. A fixed main frame body is provided with a mechanical arm system, a control system and an evaluation system; an inductive sword target is installed at the tail end of a target feeding mechanical arm; an auxiliary training sword is installed at the tail end of a sword holding mechanical arm; a motion sensor is arranged on the inductive training sword to identify the sword holding posture and sword speed of a trainer in real time; an inductive shield is provided with an inductive system to detect the defense times of the trainer; a heart rate sensor is used to identify the heart rate change of the trainer in real time; the control system comprises a data analysis module, a feedback adjustment module, a trajectory planning module and a safety protection module; and the evaluation system is used to evaluate the reaction time, sword action and heart rate of the trainer. The main body of the device is provided with the inductive sword target and the auxiliary training sword, the trainer is provided with the inductive training sword and the inductive shield, and the sensing detection and control method are combined to automatically adjust the training difficulty, so that the intelligent degree is high, the sword performance of the trainer can be optimized, and the device has the advantages of high intelligence and high training efficiency.

Description

Technical Field

[0001] This invention relates to the field of training equipment technology, specifically a swordsmanship training device. Background Technology

[0002] Swordsmanship training, as a crucial area for competitive sports and self-defense skills development, has always revolved around the accuracy of combat simulation and the scientific nature of training. Traditional training methods rely on human sparring partners and fixed targets, which inherently suffer from drawbacks such as the inability to quantify training intensity and the lack of real-time feedback on movement standardization. Although modern electronic devices have incorporated basic sensing technology to record the number of strikes and their force, it is still difficult to achieve dynamic interaction in offensive and defensive coordinated training, let alone simulate the multi-dimensional technical movements in real combat.

[0003] Current mainstream training equipment generally suffers from the following technical limitations:

[0004] Insufficient flexibility of mechanical actuators: Most mechanical devices with rigid structures have problems such as stiff movement trajectories and delayed response when simulating complex movements such as slashing and blocking. They cannot reproduce the flexibility requirements of real swordsmanship and are prone to causing joint injuries to trainees.

[0005] Data collection dimensions are too limited: Existing systems mostly focus on monitoring limb movement trajectories, neglecting the synchronous analysis of physiological indicators such as heart rate and muscle reaction speed, resulting in a lack of scientific basis for training intensity assessment and body load management;

[0006] The training mode is rigid: the fixed preset difficulty level is difficult to adapt to the advanced needs of trainees with different skill levels, especially in high-level offensive and defensive transition training, it is impossible to make dynamic strategy adjustments based on real-time performance;

[0007] Lack of equipment synergy: Training equipment for the attack and defense ends often operates independently, making it impossible to construct training scenarios that link the attack and defense rhythms, resulting in a significant difference between training effects and actual combat.

[0008] Furthermore, existing technologies offer limited support for personalized training and lack solutions that organically integrate movement standardization, physiological state monitoring, and intelligent difficulty adjustment. Particularly in simulating real combat, balancing the dynamic response speed and compliance of mechanical devices, and establishing comprehensive analytical models for multi-source heterogeneous data, have become key technological bottlenecks hindering the intelligent development of swordsmanship training equipment. Therefore, there is an urgent need for an integrated training device capable of achieving highly realistic movement simulation, multi-dimensional physiological monitoring, coordinated offensive and defensive training, and intelligent difficulty adjustment, to drive the innovation of swordsmanship training technology towards a more scientific and precise direction. Summary of the Invention

[0009] To address the lack of interactive mechanisms and poor flexibility in existing sword training equipment, this invention provides a sword training device. The main body of the device is equipped with a sensor-activated sword target and an auxiliary training sword, while the trainee is provided with a sensor-activated training sword and a sensor-activated shield. By combining sensing detection and control methods, the training difficulty can be automatically adjusted. It has a high degree of intelligence and helps to optimize the trainee's sword performance.

[0010] To achieve the above objectives, the present invention adopts the following technical solution: a sword training device, comprising a fixed main frame, a robotic arm system, a sensor-guided training sword, a sensor-guided shield, a heart rate sensor, a control system, and an evaluation system;

[0011] The fixed main frame is equipped with a robotic arm system, a control system, and an evaluation system;

[0012] The robotic arm system includes a target delivery robotic arm and a sword-holding robotic arm. The target delivery robotic arm is equipped with a sensor sword target at its end. The sensor sword target is equipped with a force sensor. The control system controls the target delivery robotic arm to drive the sensor sword target to perform the target delivery action. The sword-holding robotic arm is equipped with an auxiliary training sword at its end. The control system controls the sword-holding robotic arm to drive the auxiliary training sword to simulate swordsmanship actions.

[0013] The inductive training sword is equipped with a motion sensor that can identify the trainee's sword-holding posture and sword-drawing speed in real time.

[0014] The sensor shield is equipped with a sensing system that can detect the number of times the trainee defends.

[0015] The heart rate sensor identifies changes in the trainee's heart rate in real time.

[0016] The control system includes a data analysis module, a feedback adjustment module, a trajectory planning module, and a safety protection module. The data analysis module collects data on the force, time, and frequency of the trainee hitting the sensor-guided sword target using a force sensor, and combines this data with control data from the target-delivering robotic arm to analyze the trainee's accuracy in hitting the target with the sensor-guided training sword. The feedback adjustment module sets evaluation indicators and, based on the difference between these indicators and the accuracy obtained from the data analysis module, automatically adjusts the frequency and speed of the target-delivering robotic arm driving the sensor-guided sword target using a PID control method with neural network parameter tuning. The trajectory planning module generates the desired trajectory of the auxiliary training sword based on a preset swordplay action database using trajectory interpolation. The safety protection module includes a visual recognition system and a safety interlock system. The visual recognition system monitors the distance between the auxiliary training sword and the trainee in real time, and the safety interlock system detects the force value of the auxiliary training sword using a pressure sensor. If the distance between the auxiliary training sword and the trainee or the force value of the auxiliary training sword exceeds a warning value, the power supply is cut off via an emergency power-off switch.

[0017] The evaluation system combines data collected by motion sensors and heart rate sensors with training duration to assess the trainee's reaction time, sword techniques, and heart rate.

[0018] Furthermore, both the target delivery robotic arm and the sword-holding robotic arm are four-degree-of-freedom robotic arms, including one 60 kg·cm servo motor set at the upper arm joint and three 20 kg·cm servo motors set at the elbow joint, forearm joint and hand joint respectively. The control data issued by the control system is parsed by a CAN protocol analyzer and sent to each servo motor to realize the movement control of the robotic arm.

[0019] Furthermore, the sensing system of the sensor shield uses conductive foam in the interlayer of the sensor shield and a voltage divider circuit to convert the resistance change into a voltage signal, which is then transmitted to the control system. This identifies the number of times the sensor shield blocks the auxiliary training sword. The difference between the number of times the auxiliary training sword is swung by the robotic arm and the number of times the sensor shield blocks the sword is used to obtain the number of times the trainee successfully defends.

[0020] Furthermore, the formula for the PID control method using neural network parameter tuning is as follows:

[0021]

[0022] v target =v max (1-FI)

[0023] u(t) = K p (v target -v actual )

[0024] In the formula, FI is the fatigue index, HR is the current heart rate, and HR is the current heart rate. rest For resting heart rate, HR max =220 - age, v target v is the target delivery rate. max The maximum target delivery rate is given by u(t), where u(t) is the control variable acting on the target delivery robotic arm, and K is the value of K. p For PID parameters, v actual This represents the actual target delivery rate.

[0025] Furthermore, the assessment of reaction time involves calculating the time interval from when the target-delivering robotic arm begins delivering the sensor target to when the sensor training sword strikes the surface of the sensor target, which is taken as the trainee's reaction time.

[0026] Furthermore, the evaluation of the swordplay movements involves acquiring the trainee's sword-holding posture and sword-drawing speed through motion sensors on the training sword. A swordplay movement standard evaluation model is constructed based on a convolutional neural network (CNN). The input is time-series data of the angle and speed of the motion sensors in three directions in space, the time length is the standard execution time of each movement, and the output is a score of 0 to 1. The closer the score is to 1, the more standard the movement is.

[0027] Compared with existing technologies, the beneficial effects of this invention are as follows: The main body of the device is equipped with a sensor-guided fencing target and an auxiliary training sword, while the trainee is equipped with a sensor-guided training sword and a sensor-guided shield. It supports a training mode that coordinates fencing target practice and shield defense. By combining various sensing and control methods, the training difficulty can be automatically adjusted according to the trainee's level, thus providing a personalized training experience. Two four-degree-of-freedom robotic arms achieve high-performance, compliant control of the sensor-guided fencing target and auxiliary training sword, simulating swordsmanship movements. By comprehensively considering data from motion sensors, heart rate sensors, and training duration, the system can assess the trainee's reaction time, swordsmanship movements, and heart rate, providing comprehensive feedback. It is suitable for personalized training in various scenarios and of varying difficulty. Through real-time monitoring and feedback, it helps optimize the trainee's swordsmanship performance, providing more precise and flexible support for swordsmanship training. With a high degree of intelligence, it has significant meaning and application value for the innovation and development of swordsmanship training technology. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall structure of the sword training device of the present invention;

[0029] Figure 2 This is a system block diagram of the sword training device of the present invention.

[0030] In the diagram: 1. Fixed main frame; 2. Target delivery robotic arm; 3. Sword-holding robotic arm; 4. Sensor-activated sword target; 5. Sensor-activated training sword; 6. Sensor-activated shield. Detailed Implementation

[0031] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] like Figures 1-2 As shown, a sword training device includes a fixed main frame 1, a robotic arm system, a sensor-guided training sword 5, a sensor-guided shield 6, a heart rate sensor, a control system, and an evaluation system.

[0033] The fixed main frame 1 is equipped with a robotic arm system, a control system, and an evaluation system.

[0034] The robotic arm system includes a target-delivering robotic arm 2 and a sword-holding robotic arm 3. The target-delivering robotic arm 2 has a sensor-equipped sword target 4 at its end. The sword target 4 is equipped with a force sensor to collect the force of the trainee hitting the target. The control system controls the target-delivering robotic arm 2 to drive the sword target 4 in a target-delivering motion, assisting in guiding and detecting the trainee's sword-holding attack. The sword-holding robotic arm 3 has an auxiliary training sword at its end. The control system controls the sword-holding robotic arm 3 to drive the auxiliary training sword to simulate swordsmanship movements, assisting in guiding and practicing the trainee's shield defense. Both the target-delivering robotic arm 2 and the sword-holding robotic arm 3 are four-degree-of-freedom robotic arms, including one 60 kg·cm servo motor located at the upper arm joint and three 20 kg·cm servos located at the elbow, forearm, and hand joints respectively. Control data from the control system is parsed by a CAN protocol analyzer and sent to each servo motor to achieve motion control of the target-delivering robotic arm 2 and the sword-holding robotic arm 3. A 24V adjustable power supply is used for power supply.

[0035] The sensor-guided training sword 5 is held by the trainee, and the sensor-guided training sword 5 is equipped with a motion sensor that can identify the trainee's sword-holding posture and sword-drawing speed in real time.

[0036] The sensor shield 6 is held by the trainee. The sensor shield 6 is equipped with a sensing system that can detect whether the trainee has successfully defended against the sword techniques output by the auxiliary training sword controlled by the sword-holding robotic arm 3. Specifically, the sensing system refers to the conductive foam placed in the interlayer of the sensor shield 6. When hit by the auxiliary training sword, the resistance of the conductive foam decreases. Combined with a voltage divider circuit, the change in resistance is converted into a voltage signal and transmitted to the control system. This identifies the number of times the sensor shield 6 blocks the auxiliary training sword. By combining the number of times the auxiliary training sword is swung by the sword-holding robotic arm 3 with the number of times the sensor shield 6 blocks, the number of times the trainee successfully defends can be determined.

[0037] The heart rate sensor is worn by the trainee and can identify changes in the trainee's heart rate in real time.

[0038] The control system controls the target-feeding robotic arm 2 and the sword-holding robotic arm 3, wherein:

[0039] The control of the target-feeding robotic arm 2 can adaptively change the target-feeding rhythm, including a data analysis module and a feedback adjustment module. The data analysis module collects data on the force, time, and frequency of the trainee hitting the sensor-equipped sword target 4 using a force sensor, and combines this data with the control data of the target-feeding robotic arm 2 to analyze the trainee's accuracy in hitting the sensor-equipped sword target 4 with the sensor-equipped training sword 5. The feedback adjustment module sets evaluation indicators. Based on the difference between the evaluation indicators and the accuracy rate obtained by the data analysis module, it automatically adjusts the frequency and speed of the target-feeding robotic arm 2 driving the sensor-equipped sword target 4 using a PID control method with neural network parameter tuning. If the accuracy rate is higher than the evaluation indicator, the frequency and speed of the target-feeding are increased; conversely, the frequency and speed are decreased, achieving targeted changes in the training rhythm. The core formula of the PID control method with neural network parameter tuning is as follows:

[0040]

[0041] v target =v max (1-FI)

[0042] u(t) = K p (v target -v actual )

[0043] In the formula, FI is the fatigue index, HR is the current heart rate, and HR is the current heart rate. rest For resting heart rate, HR max =220 - age, v target v is the target delivery rate. max The maximum target delivery rate is given by u(t), where u(t) is the control variable acting on the target delivery robotic arm, and K is the value of K. p For PID parameters, v actual This represents the actual target delivery rate.

[0044] The control of the robotic arm 3 can be configured to simulate swordplay movements and implement safety braking mechanisms, including a trajectory planning module and a safety protection module. The trajectory planning module generates the desired trajectory of the training sword based on a pre-set swordplay movement database using trajectory interpolation. The safety protection module includes a visual recognition system and a safety interlock system. The visual recognition system monitors the distance between the training sword and the trainee in real time. The safety interlock system detects the force applied to the training sword using a pressure sensor. If the distance between the training sword and the trainee or the force applied to the training sword exceeds a warning threshold, an emergency power-off switch cuts off the power supply to ensure the safety of the training process.

[0045] The evaluation system combines data collected by motion sensors and heart rate sensors with training duration to assess the trainee's reaction time, sword techniques, and heart rate after training.

[0046] The reaction time is evaluated by calculating the time interval from when the target delivery robot arm 2 starts delivering the target to the trainee strikes the surface of the target with the sensor training sword 5.

[0047] The evaluation of swordplay movements is carried out by acquiring the trainee's sword-holding posture and sword-drawing speed through motion sensors on the training sword 5. A standard evaluation model for swordplay movements is constructed based on a convolutional neural network (CNN). The input is the time series data of the angle and speed of the motion sensor in three directions in space, the time length is the standard execution time of each movement, and the output is a score of 0 to 1. The closer the score is to 1, the more standard the movement is.

[0048] The heart rate assessment combines the trainee's heart rate data and training duration with a pre-set physiological model to evaluate the trainee's endurance level.

[0049] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0050] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A swordsmanship training device, characterized in that: It includes a fixed main frame (1), a robotic arm system, a sensor training sword (5), a sensor shield (6), a heart rate sensor, a control system, and an evaluation system; The fixed main frame (1) is equipped with a robotic arm system, a control system, and an evaluation system; The robotic arm system includes a target delivery robotic arm (2) and a sword-holding robotic arm (3). The target delivery robotic arm (2) is equipped with a sensor sword target (4) at its end. The sensor sword target (4) is equipped with a force sensor. The control system controls the target delivery robotic arm (2) to drive the sensor sword target (4) to perform the target delivery action. The sword-holding robotic arm (3) is equipped with an auxiliary training sword at its end. The control system controls the sword-holding robotic arm (3) to drive the auxiliary training sword to simulate sword fighting actions. The induction training sword (5) is equipped with a motion sensor that can identify the trainee’s sword-holding posture and sword-drawing speed in real time. The sensor shield (6) is equipped with a sensor system that can detect the number of times the trainee defends; The heart rate sensor identifies changes in the trainee's heart rate in real time. The control system includes a data analysis module, a feedback adjustment module, a trajectory planning module, and a safety protection module. The data analysis module collects the force, time, and frequency data of the trainee hitting the sensor sword target (4) through a force sensor, and analyzes the accuracy of the trainee hitting the sensor sword target (4) with the sensor training sword (5) by combining the control data of the target delivery robot arm (2) with the target delivery action. The feedback adjustment module sets evaluation indicators, and automatically adjusts the frequency and speed of the target delivery robot arm (2) driving the sensor sword target (4) to deliver the target based on the difference between the evaluation indicators and the accuracy obtained by the data analysis module through a PID control method with neural network parameter tuning. The trajectory planning module generates the expected trajectory of the auxiliary training sword based on a preset sword action database and using a trajectory interpolation method. The safety protection module is equipped with a visual recognition system and a safety interlock system. The visual recognition system monitors the distance between the auxiliary training sword and the trainee in real time. The safety interlock system detects the force value of the auxiliary training sword through a pressure sensor. If the distance between the auxiliary training sword and the trainee or the force value of the auxiliary training sword exceeds the warning value, the power supply is cut off through an emergency power-off switch. The evaluation system combines data collected by motion sensors and heart rate sensors with training duration to assess the trainee's reaction time, sword techniques, and heart rate.

2. The sword training device according to claim 1, characterized in that: Both the target delivery robotic arm (2) and the sword-holding robotic arm (3) are four-degree-of-freedom robotic arms, including one 60kg·cm servo motor set in the upper arm joint and three 20kg·cm servo motors set in the elbow joint, forearm joint and hand joint respectively. The control data issued by the control system is parsed by the CAN protocol analyzer and sent to each servo motor to realize the movement control of the robotic arm.

3. The swordsmanship training device according to claim 1, characterized in that: The sensing system of the sensing shield (6) uses conductive foam in the interlayer of the sensing shield (6) and a voltage divider circuit to convert the resistance change into a voltage signal and transmit it to the control system. This identifies the number of times the sensing shield (6) blocks the auxiliary training sword. The difference between the number of times the auxiliary training sword is swung by the sword-holding robotic arm (3) and the number of times the sensing shield (6) blocks the sword is obtained, thus obtaining the number of times the trainee successfully defends.

4. The sword training device according to claim 1, characterized in that: The formula for the PID control method of neural network parameter tuning is as follows: V target =v max (1-FI) u(t)=K p (v target -v actual ) In the formula, FI is the fatigue index, HR is the current heart rate, and HR is the current heart rate. rest For resting heart rate, HR max =220 - age, v target v is the target delivery rate. max The maximum target delivery rate is given by u(t), where u(t) is the control variable acting on the target delivery robotic arm, and K is the value of K. p For PID parameters, v actual This represents the actual target delivery rate.

5. The sword training device according to claim 1, characterized in that: The reaction time assessment is calculated by taking the time interval from when the target delivery robot arm (2) starts delivering the target to the moment when the training sword (5) strikes the surface of the target (4), and using this time as the trainee's reaction time.

6. The sword training device according to claim 1, characterized in that: The evaluation of the swordplay movements is carried out by obtaining the trainee's sword-holding posture and sword-drawing speed through the motion sensor on the training sword (5). A swordplay movement standard evaluation model is constructed based on the convolutional neural network (CNN). Its input is the time series data of the angle and speed of the motion sensor in three directions in space, the time length is the standard execution time of each movement, and the output is a score of 0 to 1. The closer the score is to 1, the more standard the movement is.

Images