from the Developmental Robotics Laboratory at Iowa State has started a projects to explored how proprioceptive sensory feedback, in the form of detected joint motor efforts, can be used by the PR2 robot and used by it for object perception. To allow it to tell when a bottle is either empty or full.
For the PR2 telling whether a bottle is empty or full is a tricky hurdle to complete as it cant use its cameras or 3D laser scans can help it. Watch the video after the break to see the PR2 in action.
The PR2 was put through its paces and was asked to solve:
– a sorting task in which only empty bottles are cleared off a table
– estimate the weight of bottles
– sliding boxes across a table
Within about 10-20 minutes of training the PR2 was able to learn an accurate model for distinguishing between full and empty boxes.
Proprioceptive Sensory Feedback
Proprioceptive sensory feedback is a critical aspect of robotics, particularly in tasks that require fine motor skills and object manipulation. This type of feedback allows robots to understand the position and movement of their joints and limbs without relying on external sensors like cameras or laser scanners. In the case of the PR2 robot, proprioceptive feedback is used to detect the effort exerted by its joints when lifting or moving objects. This information is then processed to determine the weight and, consequently, whether a bottle is empty or full.
The ability to distinguish between different weights is not just a party trick; it has significant implications for the future of robotics. For instance, in industrial settings, robots could use similar techniques to handle fragile items or sort products based on weight. In healthcare, robots could assist in lifting patients or moving medical supplies, ensuring that they handle each item with the appropriate amount of force.
Training and Applications
The training process for the PR2 robot involved repetitive tasks that allowed it to build a model of how different weights feel through its joints. This model is crucial for the robot to make accurate distinctions between full and empty bottles. The training was surprisingly quick, taking only about 10-20 minutes for the PR2 to develop a reliable model. This rapid learning capability is a testament to the advancements in machine learning and robotics.
Beyond sorting bottles, the PR2 was also tasked with sliding boxes across a table. This task further tested its ability to gauge weight and apply the correct amount of force. The success of these tasks demonstrates the robot’s versatility and potential for various applications. For example, in a warehouse setting, the PR2 could be used to sort packages based on weight, ensuring that heavier items are placed on sturdier shelves while lighter items are stored in more accessible locations.
Another exciting application could be in the realm of home automation. Imagine a future where robots like the PR2 can assist with household chores, such as sorting recyclables, lifting heavy grocery bags, or even helping with cooking by measuring ingredients based on weight. The possibilities are endless, and the development of proprioceptive sensory feedback is a significant step toward making these scenarios a reality.
Via Ubergizmo Via Plastic Pals
Latest Geeky Gadgets Deals
Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
