A team of computer scientists has discovered a robotic arm using artificial intelligence. The pioneering research has paved the way to pack boxes efficiently and has thus saved time and money for businesses.
A YouTube video based on the innovative invention displays a Kuka robotic arm, which is tightly packing objects from a bin and putting these into a shipping order box. The actual speed is anticipated to be five times more than the actual speed. Rutgers scientific team observed that tightly packing boxes from an unorganized pile has been one of the critical reasons for warehouse inefficiency. Thus by automating such menial tasks, the team hopes to escalate the level of the company’s competitiveness. The robotic hand invention will also reduce the physical labor of the workers.
“We can achieve low-cost, automated solutions that are easily deployable. The key is to make minimal but effective hardware choices and focus on robust algorithms and software,” said Kostas Bekris, senior author of the research.
According to reports, the team dealt with multiple aspects focusing on robot packing problem. For this, the team allegedly used hardware, robust motion and 3D perception. The team’s success is centered on the task of the robotic arm’s efficiency of placing objects from a bin into a small shipping box, followed by the task of tightly arranging them. This is attributed to be a more difficult task for a robot in comparison to just picking up boxes and dropping them elsewhere.
In order to achieve the desired results, the researchers developed a special algorithm and software for the robotic arm. With the help of visual data and a simple suction cup taking the role of a finger, the arm could push objects. Moreover, the arm uses sensor data to pull objects to a targeted area and also to push them together. For such tasks, the arm uses real-time monitoring to detect and overcome failures.
The research has been published in IEEE International Conference on Robotics and Automation. The research aims to make further advances in exploring boxes of different shapes and sizes. The team also focuses on exploring automatic learning after it is given a specific task.