As technology iterates and innovates, robots are playing an increasingly important role in human life, whether in medical devices or industrial manufacturing, where intelligent robots can be found. The key to empowering robots to perceive the world and feed back information lies in technological innovations in vision and force perception. Currently, machine vision is relatively well accomplished, supported by high-resolution imaging and mature algorithms. In the area of force perception, which is a more sophisticated test of a robot's ability to operate, the competition is not fully reflected and the market is still a blue ocean.
一. Introduction to six-dimensional force sensors
Six-dimensional force sensors are devices that use force sensitive elements and conversion elements and processing units to convert one or more mechanical quantities (tension, pull, pressure, weight, torque, internal stress and strain, etc.) into electrical signal outputs (voltage, current, bus, etc.); while six-dimensional sensors are sensors capable of measuring three force components and three moment components at the same time, and are the most complete form of multi-dimensional force sensors, currently mainly industrial robots, collaborative robots, medical robots and other applications.
The development of force sensors involves a multidisciplinary cross-fertilisation of mechanics, mechanics, materials and metrology, and the barriers are high. The main technologies determining the sensors include the following:
1、Sensitivity
When the sensitivity is high, the value of the output signal corresponding to the measured change is only relatively large, which is conducive to signal processing. The sensitivity of the sensor is high, the external noise unrelated to the measured is also easy to mix, but also be amplified by the amplification system, affecting the measurement accuracy.
The sensitivity of the sensor is directional. When the measured is a one-way quantity, and its directional requirements are high, it should choose other directional sensitivity of small sensors; if the measured is a multi-dimensional vector, the smaller the sensor cross-sensitivity required, the better.
2, frequency response
Frequency response characteristics of the sensor determines the measured frequency range, must be in the allowable frequency range to maintain no distortion. In fact, the response of the sensor always - a certain delay, the shorter the delay, the better. The higher the frequency response of the sensor, the wider the range of signal frequencies that can be measured.
3, linear range
The linear range of the sensor refers to the output and input proportional to the range. The wider the linear range of the sensor, the larger the range, and to ensure a certain degree of measurement accuracy. When the required measurement accuracy is relatively low, within a certain range, the non-linear error can be smaller sensor approximation as linear, which will bring great convenience to the measurement.
4, stability
Factors affecting the long-term stability of the sensor in addition to the structure of the sensor itself, mainly the sensor's use of the environment, the sensor must have a strong environmental adaptability. In the selection of sensors before the use of its environment should be investigated, and according to the specific use of the environment to select the appropriate sensor, or take appropriate measures to reduce the impact of the environment.
5、Accuracy
The accuracy of the sensor as long as to meet the accuracy requirements of the entire measurement system can be, do not have to choose too high. In this way, it is possible to choose a cheaper and simpler sensor among the many sensors that meet the same measurement purpose Atlas air compressor parts.
If the purpose of measurement is qualitative analysis, the use of high repeatability of the sensor can be, should not choose the absolute value of the high accuracy; if it is for quantitative analysis, must obtain accurate measurement values, you need to use the accuracy level can meet the requirements of the sensor.
二. the main application scenarios
①Industrial robot collision detection
In order to prevent damage to robots or objects due to excessive interaction during operation, flexible control is necessary to complete these tasks, so collision strength is an important part of active flexible control; on the other hand, many tasks now also require human intervention, and in terms of safety, it is necessary to detect whether there is a collision between the robot and a person, to keep the person as safe as possible, and to allow the robot to recognise the presence of a person so that the work in progress can be stopped if necessary to avoid accidentally injuring people.
②Precision grinding by industrial robots
At present, most domestic manufacturing enterprises use manual or hand-held pneumatic electric tools for grinding, lapping and filing for deburring, grinding and polishing of engine casings, bodywork and other workpieces, which can easily lead to a rise in the defective product error rate and low efficiency, and can lead to problems such as uneven surface roughness of processed products; at the same time, the sparks, dust and noise generated by grinding have a great impact on the physical and mental health of workers The quality of manual grinding needs to be roughly judged from the workers' experience, making the quality of grinding not guaranteed; the lack of skilled labour, low labour efficiency, difficulty in recruiting labour, the continuous rise in labour costs and the poor working environment have prompted the trend of robots replacing manual labour to accelerate development.
③Industrial robot dragging and teaching
Drag teaching technology can be used to teach robots in an intuitive way by directly hand-holding the robot to reach a specified position or move along a specific trajectory, while recording the position data of the teaching process, which can significantly shorten the programming efficiency of industrial robots in the application deployment stage, reduce the requirements for operators, and achieve the purpose of reducing costs and increasing efficiency.
④Medical rehabilitation
The six-dimensional force sensor with micron-level accuracy breaks through the lack or inadequacy of haptic feedback (force feedback) in medical robots, which affects the accuracy, safety and flexibility of surgical operations, and increases the robot's haptic function and sensing of touch, pressure and pull, while also simultaneously detecting the magnitude of force and the strength of haptic sensation during the medical process. This will solve the problem of surgical sequelae of medical robots due to the lack of haptic function and the lack of sensing of touch, pull and pressure, and promote the application and development of medical robots.
三. the domestic and international market situation
Six-dimensional force-controlled sensors are currently widely used in the fields of industrial fine grinding and medical rehabilitation. Although the domestic market has started late, there are many excellent companies in this segment. Compared with foreign brands, many domestic products have already met or even exceeded the relevant performance indicators of foreign products, and have more advantages in terms of price and service, therefore, there is a great opportunity to achieve domestic substitution in the future.
四. Conclusion
Force sensors give robots a sense of touch, allowing them to better perceive the world. As the application scenarios of intelligent robots continue to be enriched, six-dimensional force sensors will become an important bridge for future human-robot interaction.