A method and apparatus for encoding instructions based on the combination of wrist rotation direction and number of rotations.
By using a command encoding method that combines wrist rotation direction and number of rotations, the problem of insufficient command quantity and high false trigger rate in existing gesture interaction solutions is solved, enabling silent one-handed operation and multi-device control, and providing convenient interaction for complex devices.
Patent Information
- Application Number
- CN202610812277.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-06
- Publication Date
- 2026-11-10
AI Technical Summary
Existing gesture interaction solutions suffer from insufficient command quantity, heavy user memory burden, and high false trigger rate in controlling complex devices, while voice interaction is ineffective in certain environments.
By defining the combination of wrist rotation direction and number of rotations, using a start prefix and selection sequence encoding instructions, and combining inertial sensor recognition, multi-dimensional encoding is achieved, supporting multi-device control and emergency interruption mechanisms.
It enables silent one-handed operation, allows for exponential expansion of the number of codes, prevents accidental triggering, is suitable for controlling a variety of devices, has a low learning cost, and is suitable for noisy environments and specific scenarios.
Smart Images

Figure CN123067697A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of human-computer interaction technology, specifically relating to a method and wearable device for encoding complex instruction sets by combining wrist rotation direction and number of rotations. Background Technology
[0002] Current gesture interaction solutions generally face the problem of insufficient command quantity:
[0003] Simple gesture solutions can only recognize a few gestures (such as clenching a fist, opening, and swiping), and the number of commands that can be mapped is limited, which cannot meet the multi-level operation needs of complex devices (such as smart home systems, AR glasses, and drones).
[0004] Multi-gesture combination scheme: requires users to learn and memorize a large number of different gestures, which is a heavy memory burden, and different gestures are easily confused, resulting in a high rate of false triggering.
[0005] Voice supplementation solutions: cannot be used in noisy environments, quiet environments (such as libraries or meetings), or private scenarios, and are ineffective for special groups such as aphasic patients.
[0006] Therefore, there is a need for an interaction method that relies on a single hand gesture, can encode a large number of different instructions, and is not easily triggered by accident. Summary of the Invention
[0007] Purpose of the invention:
[0008] A command encoding method is provided, which enables users to issue a considerable number of distinguishable control commands with one hand and silently by simply combining the direction and number of wrist rotations, and can switch between controlling multiple different types of devices through encoding.
[0009] Technical solution:
[0010] A method for encoding instructions based on a combination of wrist rotation direction and number of rotations includes the following steps:
[0011] Step 1: Define Action Primitives
[0012] Based on inertial sensor data, at least two wrist rotation directions are defined as motion primitives:
[0013] The first direction (preferably inward flipping): used to indicate command prefixes or mode switching. Because this direction occurs very infrequently in daily activities, it can effectively prevent accidental triggering.
[0014] The second direction (preferably outward flipping): used to represent the main body of the instruction sequence. Because this direction's movement is natural and effortless, it is suitable for high-frequency operations.
[0015] Step 2: Constructing Encoding Rules
[0016] The instruction consists of two encoded parts: a start prefix and a selection sequence.
[0017] Start-up prefix: Defined as N consecutive (preferably 2) rapid toggle actions in the first direction. This prefix is used to wake up the system from its command listening state.
[0018] Sequence selection: After recognizing the start prefix, the system opens a preset listening time window (e.g., 3 seconds). Within this window, the system counts and / or identifies the rhythm of rapid flipping motions in the second direction.
[0019] Command Mapping: Presets the mapping relationship between different selection sequences and specific control commands. For example, flipping out once corresponds to the first command, flipping out twice corresponds to the second command, flipping out three times corresponds to the third command, long-pressing outward corresponds to the fourth command, and flipping outward twice in quick succession corresponds to the fifth command, etc.
[0020] Step 3: Switching between multi-device control modes
[0021] The control commands include entering different device control modes. Different numbers of wrist rotations correspond to activating different device control modes. For example: one outward rotation enters the mobile phone control mode, two outward rotations enter the smart home control mode, three outward rotations enter the rehabilitation exoskeleton control mode, and four outward rotations enter the drone control mode. In a certain device control mode, subsequent preset gestures (such as wrist rotations and finger movements) are mapped to the device's specific control commands. Users can switch to other device control modes by executing the activation prefix again, or exit the current mode and return to free mode by inward rotation of the wrist.
[0022] Step 4: Execution and Cancellation
[0023] Command execution: After the listening window expires or the user pauses for more than the preset interval, the system matches the corresponding control command from the preset mapping table based on the statistical number of flips or rhythm, and sends it to the target device for execution through the communication unit.
[0024] Cancellation mechanism: If the user detects the first direction (inward) flipping action again within the monitoring window, the system will immediately cancel the current instruction code, exit the monitoring state, and provide vibration feedback to confirm the cancellation.
[0025] Emergency interrupt: Regardless of the current state, if two consecutive rapid flips in any direction are detected, an emergency interrupt will be triggered unconditionally, stopping all instructions and returning to the default state.
[0026] Beneficial effects:
[0027] The number of instructions expands exponentially: Through a multi-dimensional encoding structure of "prefix + number of times + rhythm", dozens of different instructions can theoretically be encoded, far exceeding the single gesture scheme.
[0028] One-handed, silent operation: All commands can be input using only one wrist, completely freeing up the other hand and voice. Suitable for situations where both hands are occupied, noisy environments, meetings where silence is required, or private scenarios.
[0029] It has a strong ability to prevent accidental triggering: it uses "two consecutive inversions" as the start prefix, which is rarely seen in daily life, to ensure that the user's daily wrist activities (such as walking, swinging arms, picking up items, and turning over) will not trigger the command listening.
[0030] Unified control of multiple devices: Different devices can be quickly switched by flipping the wristband a certain number of times, integrating the functions of multiple remote controls into one wristband, greatly improving convenience.
[0031] Low learning cost: Users only need to remember the core rule of "first flip inward twice to start, then flip outward to select instructions", and the number of specific instructions can be gradually expanded according to proficiency. Attached Figure Description
[0033] Figure 1 : A schematic diagram of the instruction encoding format. It shows the encoding structure consisting of a "start prefix" and a "selection sequence", as well as examples of instruction mappings corresponding to different sequences.
[0034] Figure 2 Example of multi-device control mode switching mapping. It demonstrates how 1 / 2 / 3 / 4 times of eversion switch control enters the mobile phone, smart home, rehabilitation exoskeleton, and drone control modes respectively, as well as the exclusive functions of subsequent gestures in each mode.
[0035] Figure 3 Example of tiered call mapping. It shows that flipping out 1 time, 2 times, 3 times, and long press correspond to calling the first contact, the second contact, the public emergency medical system, and the silent alarm, respectively.
[0036] Figure 4 : Instruction execution flowchart. It shows the complete process from the default state → identifying the start prefix → entering the listening mode → statistically selecting the sequence → matching the instruction → executing the instruction → timeout cancellation or manual cancellation. Detailed Implementation
[0038] Example 1: Switching between multiple device control modes
[0039] The user wears an IMU bracelet. After flipping inward twice to enter listening mode:
[0040] Flipping out once = Entering "Phone Control Mode". Afterwards, flipping out again = confirm / answer, flipping inward again = go back / hang up.
[0041] Flipping out twice enters "Smart Home Mode". Afterward, flipping out again turns lights on / off, and flipping inward switches to the next device.
[0042] Three eversions = Enter "Rehabilitation Exoskeleton Mode". After that, clenching your fist = grasping, opening your fist = releasing (requires the use of finger sensors).
[0043] Four outward rolls = Entering "Drone Control Mode". Afterwards, outward rolls = Takeoff / Follow, inward rolls = Return to Home.
[0044] In any device control mode, the user can exit the current mode and return to free mode by flipping their wrist inward again. Two consecutive rapid flips in any mode will trigger an emergency interrupt.
[0045] Example 2: Tiered Emergency Call
[0046] The user is an elderly person living alone. After turning inward twice, turning outward once = calling children, turning outward twice = calling the community doctor, turning outward three times = calling 120, and long-pressing outward = silent alarm (only sending location SMS).
[0047] Example 3: Smart Home Scene Control
[0048] The user wears an IMU ring. After flipping it inward twice to enter listening mode: flipping it out once = turning the smart light on or off; flipping it out twice = adjusting the air conditioner temperature; flipping it out three times = opening or closing the curtains; flipping it out twice quickly = activating the voice assistant.
[0049] Example 4: UAV waypoint instructions
[0050] The operator wears an IMU bracelet. After flipping inward twice to enter listening mode: flipping outward once = automatic return to home; flipping outward twice = hovering and shooting a panoramic view; flipping outward three times = switching to tracking mode; long press and hold the outward flip = emergency landing.
[0051] Example 5: Quick Operation of AR / VR Virtual Menu
[0052] The user wears AR glasses and an IMU ring. After flipping inward twice to enter listening mode: flipping out once = screenshot; flipping out twice = start / stop screen recording; flipping out three times = switch display mode; flipping out quickly = bring up the quick settings panel.
Claims
1. A command encoding method based on a combination of wrist rotation direction and number of rotations, applied to a wearable control device, characterized in that, include: A preset action coding system consisting of a start prefix and a selection sequence is provided. The start prefix is defined as N consecutive rapid flipping actions in the first direction; After the startup prefix is detected, a preset listening time window is opened to count the number of rapid flipping actions towards the second direction; Based on the number of times, the corresponding control command is matched from the preset mapping relationship and executed; Wherein, the first direction and the second direction are opposite or different wrist rotation directions.
2. The method according to claim 1, characterized in that, The first direction is inward flipping, and the second direction is outward flipping.
3. The method according to claim 2, characterized in that, The start prefix is flipped inward twice consecutively.
4. The method according to claim 1, characterized in that, The mapping relationship includes: flipping once in the second direction corresponds to the first instruction, flipping twice corresponds to the second instruction, and flipping three times corresponds to the third instruction.
5. The method according to claim 4, characterized in that, The instructions are emergency call instructions or equipment control instructions at different levels.
6. The method according to claim 4, characterized in that, The control commands include entering different device control modes, where different numbers of flips correspond to activating different device control modes; in a certain device control mode, subsequent preset gestures are mapped to the device's exclusive control commands.
7. The method according to claim 1, characterized in that, It also includes a cancellation mechanism: if a flipping action in the first direction is detected again within the listening time window, the current instruction encoding is cancelled and the listening is exited.
8. The method according to claim 1, characterized in that, The selection sequence also includes the identification of the flipping rhythm, wherein short-interval continuous flipping and long-interval continuous flipping are respectively mapped to different instructions.
9. A wearable instruction encoding device for implementing the method of any one of claims 1 to 8, characterized in that, include: Inertial measurement unit: worn on the user's wrist, used to detect the direction and number of wrist rotations; Processing unit: configured to perform the start prefix identification, selection sequence counting, and instruction matching; Communication unit: used to send matched control commands to the target device; Feedback unit: Used to provide users with confirmation of the coding status.
10. An interactive system comprising the apparatus of claim 9 and a target device communicatively connected to the apparatus, wherein the target device is a smartphone, a smart home device, a drone, an exoskeleton robot, or an augmented reality / virtual reality device.
