Safety Nursing Control Method Based on Multi-sensing Food-Assisting Robot Assisted Eating

Abstract

The present invention proposes a safety nursing control method based on multi-sensing food-assisting robot to assist eating. After receiving the motion and facial features action information obtained from image recognition and motion perception data, a safety zone is set and a food-assisting robot is generated. Constrain the control strategy of the end trajectory safety zone; after receiving the information on the recognition of the mouth movement of the face and its orientation in the assisted feeding, for the three control problems in the feeding care, generate the assisted feeding safety protection control of the feeding robot Strategy: After assisted feeding, after receiving the image environment recognition detection of the plate and the patient and comparing the images before and after feeding and analyzing them, according to the requirements of feeding care, and aiming at the detection and judgment of the remaining food after feeding care and cleaning and sanitation issues, a food-assisting robot is generated Nursing effect evaluation strategy after assisted feeding. The safety nursing control strategy proposed by the invention can effectively improve the safety and nursing ability of auxiliary feeding, and avoid harm to the human body.

Description

technical field [0001] The invention relates to a safety nursing control method, in particular to a multi-sensing-based safety nursing control method for a food-assisting robot to assist eating, and belongs to the field of robot control strategies. Background technique [0002] Service robots mainly refer to robots engaged in maintenance, cleaning, rescue, guardianship and other tasks. It is a semi-autonomous or fully autonomous robot that can complete service work that is beneficial to human health. The food-assisting robot belongs to the category of service robots. The food-assisting robot is actually a robot used to help people in need eat. The food-assisting robot does not require the participation of the user. Eating and feeding a series of actions to help patients with physical disabilities to eat, complete assisted feeding, and further improve the automation and intelligence of patient care. [0003] However, in actual use, because its food-assisting robot needs to ...

AI Technical Summary

Problems solved by technology

[0004] However, the control strategies adopted by existing food-assisting robots mainly focus on the control methods or strategies for how to take food, the automatic selection of different food types, and the precise food-assisted aspects for patients. The safety protection during assisted eating is basically based on the tactile sensor triggered by the contact between the user and the robot to trigger the emergency stop of the food-assisted robot, or the emergency stop triggered by the real-time monitoring and early warning system or the guardian's external force. To achieve safety protection, but when the food-assisting robot is triggered to stop, in most cases, the patient and the robot are likely to have collided or interfered with each other, especially in some scenarios, the collision or interference has already caused serious damage to the human body. hurt

Moreover, the existing collision detection method directly determines that the entire robot system is to be safely shut down after detecting a collision or interference. It has no damage identification, judgment and evaluation capabilities for accidents or interference, and cannot adopt a higher level of safety protection. For example, in case of secondary injury or sudden disease caused by collision or mutual interference, timely alarm and simple emergency measures cannot be implemented, and the ability to eliminate collision and mutual interference is not available

At the same time, after an emergency shutdown, it takes a lot of time and manpower to recover and investigate accidents, and it is difficult to obtain specific data when accidents or accidents occur

At the same time, its existing control methods still lack the ability in nursing monitoring, and it is difficult to integrate and integrate with existing medical treatment and rehabilitation nursing.

It can be seen that the control strategies of the existing food-assisting robots in terms of safety are still quite insufficient.

Images