4results about How to "Avoid recognition errors" patented technology

The application relates to the technical field of household appliances, and discloses a cleaning brush and a control method thereof. The cleaning brush comprises a main machine, a handle and a machine head. The machine head is connected with the main machine and comprises a water tank and brush hairs. Cleaning medium is arranged in the water tank, and the brush hairs are arranged at the bottom of the water tank. An atomizing water spraying device is arranged at the bottom of the water tank and communicates with the inside of the water tank. The other end of the atomizing water spraying device is provided with a nozzle. The atomizing water spraying device is used for atomizing the cleaning medium and spraying the atomized cleaning medium to a surface to be cleaned and the brush hairs. Compared with the related art, the water tank and the atomizing water spraying device are integrated in the brush head, so that an additional handheld water pipe, a shower for humidification or frequent wetting of the brush hairs is not needed, the cleaning operation process is simplified, and the use convenience is remarkably improved.

This invention discloses a compliant control method for a massage robot with stiffness self-adaptation, comprising a six-degree-of-freedom robotic arm, an end effector, a six-dimensional force sensor, and a control host. By installing a six-dimensional force sensor at the end effector, the contact force between the robot and the human body surface can be collected in real time. Combined with end effector motion data obtained from a pose sensor, the system executes a compliant control closed loop to achieve adaptive and safe massage operation. The method can construct a force control closed loop in real time. Through the force sensor installed at the robot's end effector and combined with a compliant control algorithm based on displacement error, precise, safe, and compliant massage operation can be achieved without relying on a vision system. Simultaneously, it achieves stiffness detection and adaptive control of the massage area, effectively improving the user experience.

This invention provides a method for identifying easily confused pests based on cross-modal semantic alignment. The method includes: collecting visual data, linguistic data, and environmental data related to pests, preprocessing them, and constructing a knowledge graph of easily confused pests; constructing a modality adaptation matrix based on the visual data, text data, and environmental data; and processing the pest information currently input by the user, combining the knowledge graph and the modality adaptation matrix, and outputting the identification results of easily confused pests. This invention overcomes the shortcomings of existing single-modal and simple multimodal pest identification technologies, which have weak ability to distinguish easily confused pests and high misjudgment rates. By deeply fusing four-dimensional features (visual, linguistic, environmental, and regional) and using reasoning assistance from the knowledge graph, it achieves efficient and accurate identification of easily confused pests in different regions.

The application provides a radar working mode recognition method and device with a self-checking function, and the method comprises the following steps: using a configured shortest path method to pre-recognize the working mode of a radar state sequence; determining a mode estimation result of matching between a candidate string and a verification string based on at least one of a configured active search matching algorithm or a normal matching algorithm; and recording a mode recognition result and a corresponding sliding window moving rule. The embodiment of the application can effectively avoid recognition errors caused by improper mode division, has strong recognition ability for the anti-reconnaissance means of 'one word with multiple modes', and the method provided by the embodiment of the application does not need to create training samples for each mode, can automatically correct recognition deviation by adaptively adjusting a matching strategy through a feedback loop, improves the calculation efficiency on the premise of ensuring the recognition effect.