Wednesday 09 April 2025
Scientists have made a significant breakthrough in developing a system that can accurately estimate the weight of objects being lifted by people, using data collected from sensors embedded in their shoes. This innovation has the potential to revolutionize the design and control of upper-limb assistive exoskeletons, which are wearable robots that help people with limited mobility perform daily tasks.
The system relies on a network of 36 pressure sensors placed in the soles of specialized shoes, which measure the distribution of weight across the foot as someone lifts an object. The data is then processed using machine learning algorithms to estimate the weight being lifted. Researchers tested the system on three subjects, asking them to lift weights ranging from 2 kilograms to 10 kilograms.
The results were impressive: the system was able to accurately estimate the weight being lifted in most cases, with an average error of just 13.46%. The best-performing model, a type of artificial intelligence called Support Vector Regression (SVR), was able to achieve an accuracy rate of over 90% for weights between 3.5 kilograms and 6 kilograms.
The system has several advantages over existing methods for estimating weight. For example, it is not affected by variations in body weight or shoe fitting, which can be a problem with other systems that use sensors placed on the body or in the shoes. It also does not require complex calibration procedures, making it easier to implement and maintain.
The researchers believe that their system has the potential to improve the performance of upper-limb assistive exoskeletons, which are designed to help people with limited mobility perform daily tasks such as lifting groceries or carrying heavy objects. These devices can be particularly useful for individuals who have suffered injuries or illnesses that affect their ability to lift and move around.
In addition to its potential applications in healthcare, the system could also be used in other areas where weight estimation is important, such as in logistics or manufacturing. For example, it could help warehouse workers accurately estimate the weight of packages they are lifting, reducing the risk of injury and improving efficiency.
The development of this system demonstrates the power of combining advanced sensors with machine learning algorithms to solve complex problems. As researchers continue to refine and improve the technology, we can expect to see even more innovative applications in a wide range of fields.
Cite this article: “Unlocking Efficient Load Estimation in Exoskeletons Using Insole Sensors and Machine Learning”, The Science Archive, 2025.
Wearable Robots, Upper-Limb Assistive Exoskeletons, Pressure Sensors, Machine Learning Algorithms, Weight Estimation, Artificial Intelligence, Sensor Technology, Footwear, Mobility, Health Care
