Machines and robots always make life easier. Unlike humans, they achieve jobs with accuracy and speed. They also do not require brakes as they are never tired, resulting in companies focusing on using them as much as possible in their manufacturing processes to boost productivity. Machines and robots also help eliminate dangerous, filthy, and tedious jobs that are harmful to humans.
Although, various jobs in the working environment require human intervention, agility, resilience, and stretch, which robots cannot carry out.
The collaboration of humans and robots can create exciting opportunities for future manufacturing as it unites the best of both worlds. This association requires a close interplay between humans and robots. This association could be a tremendous advantage to expect for next-generation action.
A group of researchers from the Intelligent Automation Center at Loughborough University has published a research paper focussed on possible results for training robots to ferret out the intention of arm movement before humans articulate the movement.
According to Achim (PhD researcher), “A robot’s speed and torque need to be coordinated well because it can pose a serious threat to human health and safety”. Humans and robots should understand each other for effective teamwork, which is quite challenging as both speak different languages. They have proposed to give robots the ability to read human intentions.
The human brain generates a signal for every movement executed by the human body before its execution. Analyzing and measuring these signals can help to communicate an intention to move to a robot. Although the brain is a highly complex organ in the human body, ferreting out the pre-movement signal is challenging. The researchers focused on achieving this by confederating the frontal lobe activity of the human brain.
The researchers from Loughborough University have overcome this problem by training an AI program to detect the pre-movement pattern from an electroencephalogram (EEG). This technology allows recording human brain activity.
The research from these researchers describes the outcomes of a test carried out with eight participants. These participants had to sit in front of a computer that aimlessly produced a letter from A-Z on the screen and presses the key on the keyboard that matches the letter. The data of this experiment exhibits that the AI (artificial intelligence) system can detect when there will be a movement by a human arm up to 513 milliseconds (ms) before they move and, on average, around 300 ms before actual movement. After this, the AI system speculates which arm the participant would have moved from the EEG data, and motion sensors confirmed this intention.
The researchers tested the effect of the time advantage for a human-robot association framework in a simulation. They found that they could carry out higher efficiency and productivity for the same job using this technology concerning the conventional methods. This test was completed 8-11% faster than the regular time. The researchers also included false positives involving the EEG incorrectly communicating a human’s intention to move the robot, but the same results.
The research group aims to develop a program on this research and ultimately produce a system that can speculate where movement is directed—for example, picking a new workpiece or reaching for a tool.
In the recent finding of this research, the study would reach two things: first, this offered technology could help towards a closer, collective human-robot association that needs a considerable amount of research and engineering work to be fully established. Secondly, this can communicate that sooner than seeing robots and artificial intelligence/machine learning as a threat to human labor in manufacturing. This could also be a chance to keep the human as a vital part of the manufacturing units of the future.
As per the research group, there is a requirement to modify the essence of human work to develop a genuinely sustainable world that will not depend on exhausting physical and mental human labor.
Human-Robot Collaboration (HRC) is establishing to modify the factory shop-floor, yet there is a need for significant association between humans and robots.
Realistic HRC will create a metamorphose effect on labor productivity, quality of job, and robustness. This will exhibit a more secure and feasible labor market, overcoming physical drawbacks caused by gender, age, or disability.
Using Artificial Intelligence and EEG, this research leads us one step closer to genuine HRC.
The post An AI System Trained By Loughborough University Researchers Recognizes The Pre-Movement Patterns From An EEG appeared first on MarkTechPost.