The invention relates to a training apparatus and to a method of using a training apparatus.

It is known to use a free-hanging boxing bag to measure collisions resulting from tackling, punching and kicking the boxing bag.

According to a first aspect of the invention, there is provided a training apparatus comprising: a training implement for receiving an impact; a plurality of acceleration sensors arranged inside the training implement, each acceleration sensor configured to, in use, measure an acceleration of the training implement in response to an impact against the training implement; and a processing circuit programmed to process the measured acceleration by each acceleration sensor so as to compute information about the impact against the training implement, wherein the computed information includes an orientation of the impact against the training implement.

The impact (or collision) may be in the form of a punch with a fist, a kick using any part of a leg, or a slap, hit, strike or tackle using any other part of the human body. The training apparatus of the invention is applicable to the measurement of a single impact or a plurality of impacts against the training implement. The training apparatus may be used by, for example, a member of the public, a recreational gym goer, an expert practitioner or a professional athlete.

Limitations in conventional methods of measuring collisions against a training implement restricted the user to using a specific technique for applying the impact to the training implement and/or hitting a designated location on the training implement. Such conventional methods did not include or require the determination of the orientation (or direction) of the impact against the training implement. Whilst it may be possible to use external measurement equipment (e.g. an optical instrument) to determine the orientation of the impact against the training implement, the external measurement equipment adds costs and takes up valuable space.

In contrast, the training apparatus of the invention advantageously provides the necessary instrumentation to obtain the orientation information, either alone or together with other information about the impact against the training implement. There is no requirement for external measurement equipment to determine the orientation of the impact against the training implement. The determination of the orientation information enables a detailed characterisation of the impact against the training implement. For example, the impact may be oriented to collide against a front surface, rear surface, left side surface or right side surface of the training implement, or a surface of the training implement that is at an angle to its front surface, rear surface, left side surface or right side surface where the angle is more than 0° and less than 90°, relative to a user who is facing the training implement. It will be understood that the positions of the front, rear, left side and right side surfaces of the training implement may be fixed or may be defined relative to the user facing the training implement. Therefore, the orientation information may be used to determine a three- dimensional location of the impact on the surface of the training implement. The detailed characterisation provides the user with more complete feedback that closely resembles the actual impact, or the actual workout of which the impact is a part. In particular, the orientation information is useful for training implements that rely on unconstrained striking action and for training implements that move during multistriking to improve measurement accuracy.

The processing circuit may include a plurality of counters. Each counter may be associated with a respective one of the plurality of acceleration sensors. Each counter may be programmed to track how many times a timestamp reset has taken place with respect to the measured acceleration received by the processing circuit from the corresponding acceleration sensor.

The processing circuit may include a bandpass filter that is programmed to apply a frequency filter to the measured acceleration by each acceleration sensor. The frequency filter may be configured to remove noise from the measured acceleration by each acceleration sensor. The bandpass filter may include a programmable frequency threshold.

As mentioned above, the processing circuit may be programmed to process the measured acceleration by each acceleration sensor so as to compute other information about the impact against the training implement in addition to the orientation of the impact against the training implement. For example, the computed information may include: a force of the impact against the training implement; a location of the impact against the training implement; a timing of the impact against the training implement; a time interval between consecutive impacts against the training implement; a frequency of a plurality of impacts against the training implement; a precision of the impact against the training implement; and/or a reaction time of the impact against the training implement.

The training implement may include a trigger zone for receiving an impact to initiate a function of the training apparatus. The processing circuit may be programmed to initiate the function of the training apparatus when the location of the impact against the training implement matches a location of the trigger zone. The function may be a pre-programmed workout regime.

When the computed information includes a precision of the impact against the training implement, the computed information may include a comparison between a location of the impact against the training implement and a location of a prior impact against the training implement. Preferably the computed information includes a comparison between locations of two or more consecutive impacts against the training implement.

When the computed information includes a reaction time of the impact against the training implement, the training apparatus may include a stimulation device configured to selectively initiate a stimulus (e.g. a visual, aural or mechanical stimulus), and the reaction time of the impact against the training implement may be equal to the time between the initiation of the stimulus and the impact against the training implement. Optionally the stimulation device may be configured to selectively initiate the stimulus (e.g. a visual or mechanical stimulus) to provide a target on the training implement for the impact against the training implement, where the reaction time of the impact against the training implement may be equal to the time between the initiation of the stimulus and the impact against the target on the training implement.

In a preferred embodiment of the invention, the processing circuit includes a processor and memory including computer program code, the memory and computer program code configured to, with the processor, enable the processing circuit at least to process the measured acceleration by each acceleration sensor so as to compute the information about the impact against the training implement. The processing circuit may be, may include, may communicate with or may form part of one or more of an electronic device, a portable electronic device, a portable telecommunications device, a microprocessor, a mobile phone, a personal digital assistant, a tablet, a phablet, a desktop computer, a laptop computer, a server, a cloud computing network, a smartphone, a smartwatch, smart eyewear, and a module for one or more of the same. Optionally the training apparatus may include a feedback device, such as a display device (e.g. a touchscreen, a television, a monitor, a projector or a beamer) and/or an audio device. The feedback device may be configured to receive the computed information from the processing circuit. The feedback device may be configured to, in use, provide the received computed information as feedback. The display device may include a display screen for displaying the received computed information. The audio device may include a speaker for transmitting the received computed information as sound. The feedback device may be, may include, may communicate with or may form part of one or more of another feedback device, another display device (e.g. a touchscreen, a television, a monitor, a projector or a beamer), another audio device, an electronic device, a portable electronic device, a portable telecommunications device, a mobile phone, a personal digital assistant, a tablet, a phablet, a desktop computer, a laptop computer, a smartphone, a smartwatch, smart eyewear, a monitor, a television, and a module for one or more of the same. The processing circuit and the feedback device may be part of the same electronic equipment, or different electronic equipment.

The feedback device may be configured to provide the same information to multiple users, or provide different information to different users or user groups. For example, a display device may use a split-screen format, or may have multiple display screens, to show different information to different users or user groups. This may be used to create challenges for different users or user groups and/or create competition between the different users or user groups.

Further optionally the processing circuit and feedback device may be coordinated to enable feedback of the received computed information in real-time. This permits the user to receive near-instantaneous feedback on the impact, or the actual workout of which the impact is a part, which helps the user to make real-time adjustments to their training.

Alternatively the computed information may be stored in a memory of or associated with the processing circuit, the feedback device (e.g. a display device and/or an audio device) or another device, so that the computed information may be fed back to a user at a later time.

In embodiments of the invention, the training apparatus may include a hollow tube inserted inside the training implement, wherein the acceleration sensors are mounted inside the hollow tube. The hollow tube may be, for example, a pipe. In further embodiments of the invention, the training implement may be filled with a gravity-resistant filling material. For example, the gravity-resistant filling material may include a plurality of filling material portions arranged to form air pockets between the filling material portions, where the filling material portions are sufficiently coarse to prevent packing of the material portions that reduces or removes the air pockets. This helps to maintain an even distribution of the filling material along the height of the training implement over time, thus maintaining its shape, which in turn allows the training apparatus to provide more accurate and consistent information about the impact against the training implement over longer periods of time.

In contrast, when a training implement is filled with a non-gravity-resistant filling material, gravity will tend to pull the filling material downwards and thereby change the distribution of the filling material along the height of the training implement over time, thus causing the training implement to become heavier and harder towards and at its bottom end than towards and at its top end in comparison to the original state of the training implement. For example, when the gravity-resistant filling material includes a plurality of filling material portions arranged to form air pockets between the filling material portions, use of fine filling material portions would result in gravity- induced packing of the material portions that reduces or removes the air pockets. The redistribution of the non-gravity-resistant filling material along the height of the training implement may adversely influence the information about the impact against the training implement that are taken over long periods of time.

In still further embodiments of the invention, the training implement may be filled with a foam filling material that is arranged to fix a position of each acceleration sensor inside the training implement. In particular, the training implement may be filled with high-density foam filling material. This provides a cost-effective and efficient way of fixing the position of the acceleration sensors inside the training implement in order to ensure consistent and reliable acceleration measurements.

The training implement may be, but is not limited to, a training bag. The training implement may be arranged as a free-hanging training implement.

The training implement may be arranged as a standing training implement. For example, the training implement may be a standing dummy. The acceleration sensors may vary in type and number. Each acceleration sensor may be or may include an accelerometer. In a preferred embodiment of the invention, each accelerometer may be a biaxial or triaxial accelerometer.

According to a second aspect of the invention, there is provided a method of using a training apparatus according to any one of the first aspect of the invention and its embodiments, the method comprising the steps of: by each acceleration sensor, measuring an acceleration of the training implement in response to an impact against the training implement; and by the processing circuit, processing the measured acceleration by each acceleration sensor so as to compute information about the impact against the training implement, wherein the computed information includes an orientation of the impact against the training implement.

The features and advantages of the first aspect of the invention and its embodiments apply mutatis mutandis to the features and advantages of the second aspect of the invention and its embodiments.

It will be appreciated that the use of the terms "first" and "second", and the like, in this patent specification is merely intended to help distinguish between similar features, and is not intended to indicate the relative importance of one feature over another feature, unless otherwise specified.

Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, and the claims and/or the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and all features of any embodiment can be combined in any way and/or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and/or incorporate any feature of any other claim although not originally claimed in that manner.

Preferred embodiments of the invention will now be described, by way of non-limiting examples, with reference to the accompanying drawings in which:

Figure 1 shows a training apparatus according to an embodiment of the invention; and Figure 2 shows an acceleration of a training implement of the training apparatus of Figure 1 when an impact is applied against the training implement.

The figures are not necessarily to scale, and certain features and certain views of the figures may be shown exaggerated in scale or in schematic form in the interests of clarity and conciseness.

The following embodiments of the invention are described with reference to a training implement in the form of a free-hanging boxing bag. It will be appreciated that the following embodiments of the invention apply mutatis mutandis to other types of training implements, such as standing dummies.

Free-hanging boxing bags are used as research tools for accurately measuring collisions in tackling, punching and kicking. The advantages of using a free-hanging boxing bag are that it is strikeable at different heights (e.g. top, middle or bottom), at different orientations (e.g. front, left, right, rear or any point around the circumference of the boxing bag at a given height) and in different ways (e.g. punches, kicks or tackles) and provide a safe and practical way for users to generate collisions at varying levels of strength and effort that facilitate investigation of the user's skills and abilities.

Conventional systems, like wall mounted devices, only measure the impact that is applied to the system in a single fixed direction or at a flat/small target area, and typically use only a single accelerometer or force transducer. Such conventional systems restrict the collisions to the single fixed direction and/or the flat/small target area, limit the striking to only a small subset of actions and thereby exclude some striking techniques, and makes it difficult for a user to perform an actual workout or training programme. Consequently the investigation of the user's skills and abilities is also limited.

The invention allows for accurate measurement of force, location, orientation (i.e. direction) and frequency of impacts against any location of a boxing bag, which is outside the capabilities of the conventional systems. The measurement is done by utilising at least two accelerometers fixed to/in the boxing bag. An algorithm has been developed by the inventors to compute the force, location and orientation from the difference in the acceleration values measured by the accelerometers. This in turn enables accurate feedback to the user about their workout or training programme, including real-time tracking of their progress during the workout or training programme, which then allows them to near-instantaneously study and improve their technique.

A training apparatus according to an embodiment of the invention is shown in Figure 1 and is designated generally by the reference numeral 20. The training apparatus comprises a training implement, a measurement instrument and a processing circuit.

The training implement includes a boxing bag 22 that is made out of a canvas material and is filled with a filling material 24. In use, the boxing bag 22 is suspended using a hanging chain 26 from a support structure 28 (e.g. a support beam) or a ceiling so that the boxing bag 22 is free-hanging.

The measurement instrument includes a hollow aluminum pipe 30 that is inserted into the canvas material. This is followed by filling the canvas material with the filling material 24 around the pipe 30. The pipe 30 is arranged to extend along a central longitudinal axis of the boxing bag 22 with respect to the height of the suspended boxing bag 22.

The measurement instrument further includes a pair of triaxial accelerometers 36, each of which is located inside a respective end of the pipe 30. More specifically, each triaxial accelerometer 36 is attached to a respective lid (or cap) that is fitted to a respective end of the pipe 30, so that each triaxial accelerometer 36 is located inside the pipe 30. It will be appreciated that other means of mounting the triaxial accelerometers 36 into or onto the pipe 30 are possible, such as adhesive and fasteners. Each triaxial accelerometer 36 is exemplarily a 3-axis accelerometer, ADXL 377 (Analog Devices) 3-Axis ± 200g, Vs = 3V, nonlinearity ± 0.5%, cross-axis sensitivity ± 1.4%. Other configurations of accelerometers, such as digital accelerometers, may be used in other embodiments.

The above configuration of the measurement instrument results in an arrangement in which the triaxial accelerometers 36 are spaced apart at a known distance along a central longitudinal axis of the boxing bag 22 and thereby spaced apart at a known distance along the height of the boxing bag 22. In this way, when an impact is applied to the boxing bag 22, each triaxial accelerometer 36 is configured to, in use, measure an acceleration of the boxing bag 22 in response to the impact.

The filling material 24 may be a gravity-resistant filling material and/or a high-density foam filling material. A gravity-resistant filling material helps to maintain an even distribution of the filling material along the height of the boxing bag 22 over time. A high density foam filling material helps to fix the position of the pipe 30, preferably the top and bottom ends of the pipe 30, to prevent its movement, and thereby fix the positions of the triaxial accelerometers 36, relative to the rest of the boxing bag 22. This ensures consistent and reliable acceleration measurements over long periods of time.

The processing circuit includes a processor and memory including computer program code. The memory and computer program code are configured to, with the processor, enable the processing circuit to carry out various processing functions. In the embodiment shown, the processing circuit includes a microchip (such as an ESP microchip) and a cloud computing network. The microchip includes an antenna for wirelessly transmitting information to the cloud computing network. Preferably the microchip is located inside the pipe 30 so that the antenna extends out of the top of the pipe 30. The information saved in the cloud computing network can then be transmitted to a tablet, a display screen or any other display or audio device. In other embodiments, the processing circuit may be, may include or may form part of one or more of an electronic device, a portable electronic device, a portable telecommunications device, a microprocessor, a mobile phone, a personal digital assistant, a tablet, a phablet, a notebook computer, a laptop computer, a server, a smartphone, a smartwatch, smart eyewear, and a module for one or more of the same. It will be appreciated that references to a memory or a processor may encompass a plurality of memories or processors.

In other embodiments of the invention, the triaxial accelerometers 36 may be configured to be in communication with respective wireless transmitters, such as WiFi transmitters, that are capable of transmitting information wirelessly to a wireless receiver of a processing circuit external to the boxing bag 22.

The processing circuit is programmed to acquire the acceleration measurements from the triaxial accelerometers 36 and process the measured acceleration by each accelerometer 36 so as to compute information about an impact against the boxing bag 22. Computation of the information about the impact against the boxing bag 22 may be performed in the microchip and/or the cloud computing network. The computed information includes: a force of the impact against the boxing bag 22; a location of the impact against the boxing bag 22; an orientation of the impact against the boxing bag 22; a timing of the impact against the boxing bag 22; a time interval between consecutive impacts against the boxing bag 22; and a frequency of a plurality of impacts against the boxing bag 22. The timing of the or each impact against the boxing bag 22 may be recorded by registering the real computer time.

By processing the measurements in the microchip instead of the cloud computing network, the amount of data transmitted from the microchip to the cloud computing network is reduced, thus increasing battery lifetime when the boxing bag is not in use. This in turns allows for multiple boxing bags 22 to be connected to the same cloud computing network.

The computed information may include a precision of the impact against the boxing bag 22. For example, the location of a first impact against the boxing bag 22 may be used as a calibration standard, e.g. a three-dimensional calibration standard, and the location of the or each subsequent impact against the boxing bag 22 is compared against the location of the first impact against the boxing bag 22 to determine the precision of the or each subsequent impact against the boxing bag 22. For example, if a first strike (e.g. a jab) is at a point X, a next strike (e.g. a hook) is expected to be placed at a point Y relative to the point X. In this way, it is possible to assess the precision of a combination of strikes and also generate a heat map in the form of a human body to see where the strikes had the most impact. This may include determining the impact or force of an individual strike, or may include determining the total impact or force of a combination of strikes to assess its performance. This can also help in finding patterns, e.g. using machine learning, in order to identify suitable or impactful combinations for a particular user.

The computed information may include a reaction time of the impact against the boxing bag 22. For example, the training apparatus 20 may be designed to provide a stimulus, such as a visual stimulus using a light source, an aural stimulus using a sound source or a mechanical stimulus using a mechanical actuator. The reaction time of the impact against the boxing bag 22 is equal to the time between the initiation of the stimulus and the impact against the boxing bag 22. The light source may be internal or external to the boxing bag 22. The sound source may be a speaker. The mechanical actuator may be a vibrating actuator. The visual or mechanical stimulus may be configured to provide a target on the boxing bag 22 for the user to hit, where the reaction time of the impact against the boxing bag 22 is equal to the time between the initiation of the stimulus and the hitting of the target. An exemplary algorithm has been developed by the inventors for computing the force, location and orientation from the difference in the acceleration values measured by the accelerometers 36 and is described as follows, with reference to Figure 2.

The triaxial accelerometers 36 define a coordinate system that is attached to the bag. The z-axis is parallel to the central longitudinal axis of the cylindrical boxing bag 22. The coordinate system of the triaxial accelerometers 36 moves with the boxing bag 22 and has no inertial frame. The exerted force of the impact is assumed to be largely perpendicular to the surface of the bag because a hand or leg is expected to glide over the surface of the boxing bag 22 in other directions.

The algorithm employs a trigger system to use a small part of the continuously monitored accelerations around a peak of an impact. The function filters the data with a bandpass filter 20-75Hz. The low-pass 75 Hz filter removes noise from electronics and the mechanical suspension system. The high-pass 20 Hz filter removes the low frequency swinging movement of the boxing bag 22 resulting from previous strikes. The absolute values for both the top and bottom accelerometers 36, are compared to a threshold value. If at least one of the absolute values exceeds the threshold value, then a plurality of points of the acceleration data are selected and the maximum values of the selected points are identified. The highest of the maximum values is determined as the peak of the impact. The position of the peak of the impact and the selected points are then further analysed.

In the general case, the acceleration a _cm of the centre of mass is first calculated as follows: a _cm = (a _top + a _bottom )/2

Where a _top is the acceleration signal recorded by the top accelerometer 36 and a _bottom is the acceleration signal recorded by the bottom accelerometer 36 of the boxing bag 22.

The orientation φ (t) of the strike is obtained from the measured x and y components of the acceleration as follows: This is followed by a coordination rotation of the x and y acceleration component signals a _x , a _y in the dominant direction of the strike using a matrix M as follows:

Where <p is the angle at the peak of the impact on the boxing bag 22.

The resulting vector yields two signals that give the acceleration a _φ for the accelerometer 36 in the orientation of the strike and also the acceleration a _L perpendicular to the orientation of the strike. The former value is primarily used to evaluate the characteristics of the strike, while the latter value is primarily used as a comparative value to evaluate the quality of the measurements because it should be ideally zero or as close to zero as possible and much smaller than the former value. The angle y(t) is a function of time due to the rotation of the boxing bag 22 along its central longitudinal axis. The matrix M is not calculated as a function of time but is calculated for the angle at peak impact force.

The foregoing computation is carried out for the top and bottom accelerometers 36.

In the next section, all forces and accelerations are the values at the time of peak impact, t _peak , especially:

(Equation 1) a = a _φ (t _peak )

The forces acting on a free-hanging bag when impacted include the hang force, gravitational force and the reaction force due to the impact (i.e. the impact reaction force). In obtaining the impact reaction force and its location with respect to the bag's surface using the accelerometers 36, it is assumed that at the time of the impact the hang and gravitational forces are counterbalancing, which means that only the horizontal components of the accelerometers 36 are taken into account.

Under these assumptions, Equation 1 describes the equilibrium of forces acting on the boxing bag 22.

(Equation 2) F _r + ma _cm = 0

Where F _r is the bag's impact reaction force, m is the bag's mass and a _cm is the acceleration of the bag's centre of mass. To obtain the height of the impact, the angular acceleration of rotation about the horizontal axis ε _h is needed. The equilibrium of the moments of forces acting on the boxing bag 22 are given in Equation 3:

(Equation 3) F _r r _z + ε _h J _h = 0

Where ε _h is the angular acceleration of the bag's body against the centre of mass in the horizontal axis, r _z is the vertical distance of the impact reaction force to the bag's centre of mass and j _h is the moment of inertia against the bag's centre of mass.

The calculation of ε _h resembles that of the translational acceleration. The rotational vector as a function of time follows from:

(Equation 4) ε _h = (a _top + a _bottom )/L

Where L is the distance between the top and bottom accelerometers 36.

The x and y components of the rotational vector are then translated by the same matrix M. ε _h is then the value of the first component at t _peak .

The location along the central longitudinal axis relative to the bag's centre of mass at which the impact reaction force r _z is applied can be found using Equation 5:

(Equation 5) r _z = J _h ε _h / F _r

The above calculations assume that the boxing bag 22 is a rigid body. In reality the boxing bag 22 behaves as if its mass depends on the height of the impact. When the impact is at the top or bottom of the boxing bag 22, the mass of the boxing bag 22 is identical or similar to its actual mass. However, when the impact is around the middle of the boxing bag 22, the mass of the boxing bag 22 appears to be lighter than its actual mass. The inventors have observed that strikes in the middle of the boxing bag 22 results in larger bag movement than is expected from the mass of the boxing bag 22 and the force of the strike. Hence, a correction is applied by measuring the effective mass as a function of the height of the impact and fitting a polynomial to correct the measured value of the force of the impact using the effective mass, m _effective , instead of the actual mass, m. The training apparatus 20 of the invention therefore provides the necessary instrumentation to measure spatial and temporal properties of strikes to the boxing bag 22, including impact force, location, orientation and frequency. This can be used to provide analysis and feedback of a user's striking that closely resembles the actual impact, or the actual workout of which the impact is a part. In particular, the orientation and height information is especially useful in determining a three- dimensional location of the impact on the surface of the boxing bag 22, which is invaluable due to the reliance on unconstrained striking action that may be in any direction and due to the tendency of the boxing bag 22 to move during multi-striking.

An exemplary method of using the training apparatus 20 is described as follows.

The training apparatus 20 is set up as described above prior to a user's workout. The user then carries out a workout regime that may involve punching, kicking, tackling or a combination thereof that results in one or more impacts against the boxing bag 22. The workout regime may be a fitness or training programme. Each impact may be oriented to collide against a front surface, rear surface, left side surface or right side surface of the boxing bag 22, or a surface of the boxing bag 22 that is at an angle to its front surface, rear surface, left side surface or right side surface where the angle is more than 0° and less than 90°, relative to the user who is facing the boxing bag 22.

The computed information about the workout regime may be displayed in real-time on the tablet or display screen to provide the user with near-instantaneous feedback on each impact. The computed information may include details about one or more of the strength, location, height, direction, frequency, precision and reaction time of the user's strikes, which helps the user to review their workout regime and make real-time adjustments/corrections, such as increasing striking strength, changing striking technique, changing stance or posture and changing the striking frequency. The computed information may include strike intensity and/or calories burnt, which can be calculated using the aforementioned information about the user's strikes. As the user makes the adjustments/corrections, the near-instantaneous feedback by the training apparatus 20 enables the user to determine whether the adjustments/corrections have been executed properly. Also, an observer, such as a coach or researcher, may review the feedback and provide instructions to the user who is focusing on the workout regime. Alternatively the computed information may be stored in the memory of the processing circuit for display or analysis at a later time. The displayed information may additionally include one or more of: graphical visualisation of the three-dimensional location of each impact against the boxing bag 22; comparison against target values for each impact; peak performance of the user; performance of the user over time; and triggerable messages that, depending on the user's performance, provide instructions to the user to make adjustments/corrections. The target values and triggerable messages may form part of a pre-programmed workout regime that is saved in the processing circuit. The pre-programmed workout regime may be adjusted to take into account inputted values based on the user's characteristics and desired workout regime.

Additionally or alternatively the computed information about the workout regime may be fed back using an audio device, such as a speaker.

The displayed information may include the physiological zone in which the user is currently training or may include a timeline of physiological zones in which the user has trained. Examples of physiological zones include, but are not limited to, aerobic, anaerobic, recovery and heat rate zones. The ability of the invention to accurately measure impact force and measure the number of strikes per second, without requiring any spatial constraints such as the requirement to hit the boxing bag 22 at only specific locations or the requirement for the boxing bag 22 to be completely still when being struck, enables accurate identification of the physiological zone(s). When used in combination with the user's personal profile data (e.g. age, weight, gender), the displayed information can be used to indicate training load to prevent undertraining and overtraining. The displayed information can include, or be combined with, heart rate information from heart rate monitors.

The computed information may be uploaded to a database of a cloud computing network or an online platform that is accessible by a user community for a range of purposes, such as research and competitions. The database may be accessed by a user using a mobile phone or other portable computing device.

The training apparatus 20 may be configured to permit its use by multiple users, each of which has their own account programmed into the processing circuit. This enables each user to follow their own workout regime (which may be personalised) and independently monitor their own performance.

Multiple training apparatus 20 may be used, either concurrently or non-concurrently, by multiple users. The multiple training apparatus 20 may be in the same location (e.g. the same gym or home) or may be in different locations (e.g. different gyms or homes). The processing circuits of the multiple training apparatus 20 may communicate with each other and/or may communicate with a central processing circuit. Such communication preferably takes place as wireless communication, e.g. via Wi-Fi, a local server and/or the internet. This setup enables group training in which different users carry out the same workout regime or different workout regimes. This setup also enables a leaderboard system to encourage users to keep track of their performances and compete with each other. For example, data from the multiple training apparatus 20 may be sent to a single interface as part of an 'online' multiplayer mode.

The processing circuit of the training apparatus 20 may be programmed to be used with different software, e.g. a training software that can be used only during the workout, and an app that stores personal data collected by the training apparatus 20 and links to a training community. This allows personal data to be saved on a user's personal account and/or portable electronic device, e.g. a laptop or smartphone, so that they can access their personal data in various locations and can log into their account when training with different training apparatus in different locations.

The ability of the training apparatus 20 of the invention to provide complete feedback resembling the actual workout regime not only permits the development of training programmes that provide accurate feedback to professional athletes, such as combat sport athletes, and health and fitness programmes for gym users but also enables the development of pre-recorded or pre-programmed workout regimes, preferably with a gamification element, that promotes user participation and adherence to workout goals.

The training apparatus 20 may include a pre-recorded workout regime by a trainer (or trainers) that shows the user how to perform techniques and what challenges the user must perform. The pre-recorded workout regime may be delivered to the user in audio format or video format. The range of data measurable by the training apparatus 20 of the invention enables a wide range of techniques and challenges that can be performed by the user (e.g. strike the boxing bag 22 with a specific force or within a specific force range, strike the boxing bag 22 a specific number of times within a pre-defined period of time, train a specific number of minutes in a specific physiological zone, strike a specific zone of the boxing bag 22 a specific number of times, perform a specific number of full combinations within a pre-defined period of time, or a combination thereof). An example of a gamified workout regime involves the use of the impact's location and orientation information together with the boxing bag 22. The boxing bag 22 can be divided into a plurality of parts, e.g. n equal or unequal parts in the vertical direction of the boxing bag 22, the horizontal direction of the boxing bag 22 and/or around the outer surface of the boxing bag 22. The location and orientation information of an impact is used to determine which part of the boxing bag 22 has been struck by the user. This enables the boxing bag 22 to be used as a controller when the correct trigger zone is struck. For example, the user strikes a designated part of the boxing bag 22 to start the workout regime, and strikes another designated part of the boxing bag 22 to end the workout regime. This functionality can also be used to implement the boxing bag 22 as a controller for video games (such as virtual reality games, especially those involving hitting targets), where the user is required to strike the correct parts (or trigger zones) of the boxing bag 22 to progress in the game, e.g. by controlling one or more elements in the game, thereby adding a further gamification element to the workout regime. For example, a user may strike a trigger zone, e.g. the bottom, of the boxing bag 22 to start a round lasting a fixed period of time, or strike another trigger zone, e.g. the top, of the boxing bag 22 to start a round lasting a different fixed period of time. This is particularly because the orientation information provided by the invention enables the surface of the boxing bag 22 to act as a three- dimensional controller to be used in a way that is natural to the user, as opposed to a flat two-dimensional controller seen in conventional video games (such as dance pads) which comparatively has limited controller functionality. This also obviates the need for controllers based on touch screens which can be hard to press using boxing gloves.

When the boxing bag 22 is in use, the boxing bag 22 may make unexpected movements, which can cause noise to appear in the measurements from the accelerometers 36. These unexpected movements include, but are not limited to, sudden lifting and subsequent drop of the boxing bag 22, jerking of the boxing bag 22 due to its connection to the hanging chain 26, and whip-like movements of the boxing bag 22 by being struck at the bottom followed by a strike at the top.

It was observed by the inventors that the frequency range of the measured acceleration caused by a user's impact on the boxing bag 22 is distinct from the frequency ranges of the measured acceleration caused by various other movements of the boxing bag 22 which could be erroneously registered as strikes by the processing circuit. By applying a bandpass filter to the measurements by filtering out the other frequencies and thereby the associated noise, the remaining signal from the measured acceleration takes the form of a distinct spike. The bandpass filter may include a programmable frequency threshold to calibrate or fine-tune the sensitivity of the boxing bag 22. The frequency threshold may be adjusted to match the properties of the boxing bag 22, which may include its material composition, dimensions and weight, and/or match the user's profile, which may include the user's physical strength, the user's bodily dimensions, the user's physiological characteristics and the user's designated workout regime.

Implementation of the bandpass filter may impose certain restrictions such as:

• The duration of the data received from the boxing bag 22 may be limited;

• When a strike is detected, the corresponding stored data may be removed up to the point where the filtered data is below the set frequency threshold;

• While a measurement is in progress, no other measurements can be made.

Furthermore, the measurements from the accelerometers 36 may not be simultaneously received by the processing circuit. This results in a lack of synchronisation between the measurements of the accelerometers 36, which in turn adversely impacts the accuracy of the computed information.

One way of overcoming the lack of synchronisation issue is by using timestamps in the measurements from the accelerometers 36. However, due to hardware limitations, a timestamp cycle reset (overflow) may occur with respect to the measurements. For example, the measurements may be batched into groups of 100 and the timestamps may be configured as an unsigned 16-bit integer, resulting in a reset (overflow) of the timestamp cycle every minute.

To solve this issue, the processing circuit may include a plurality of counters. Each counter is associated with a respective one of the plurality of accelerometers 36. Each counter is programmed to track how many times a timestamp cycle reset has taken place with respect to the measured acceleration received by the processing circuit from the corresponding accelerometer 36. By tracking the number of timestamp cycle resets, the timing of the measurements of the accelerometers 36 can be properly synchronised to ensure the accuracy of the computed information.

The timestamp runs in cycles, and may arrive at irregular times and/or with gaps. It is therefore possible that, under certain circumstances, the wrong timestamp may be processed, leading to incorrect synchronisation between measurements of different events. To resolve this, the processing circuit may be programmed with the assumption that a given timestamp will not arrive later than a fixed period of time that is a fraction of MAX_TIMESTAMP, where MAX_TIMESTAMP is the maximum duration of the timestamp cycle before it resets.

As an example, it is assumed that a given timestamp will not arrive later than 1/n MAX_TIMESTAMP. Therefore, in this case, the timestamp cycle is split into n sections, and so timestamps in a given section is assumed to correspond to measurements of the same event. This enables the measurements of the same event to be correctly identified for synchronisation with each other, instead of measurements of different events being incorrectly identified for synchronisation with each other due to incorrect selection of timestamps.

Linear interpolation can be used to resolve any remaining lack of synchronisation between the measurements of the accelerometers 36.

It will be appreciated that the above numerical values are merely intended to help illustrate the working of the invention and may vary depending on the requirements of the training apparatus.

The listing or discussion of an apparently prior-published document or apparently prior- published information in this specification should not necessarily be taken as an acknowledgement that the document or information is part of the state of the art or is common general knowledge.

Preferences and options for a given aspect, feature or parameter of the invention should, unless the context indicates otherwise, be regarded as having been disclosed in combination with any and all preferences and options for all other aspects, features and parameters of the invention.