Home > Tag Archives: AI

Tag Archives: AI

Ditto: Building Digital Twins of Articulated Objects from Interaction

As modern AI models become larger, lots of realistic training data are needed. A recent paper published on arXiv.org introduces Ditto (Digital twin of articulated objects), an implicit neural representation-based model that jointly predicts part-level geometry and kinematic articulation between the parts. Augmented reality application example. Image credit: OyundariZorigtbaatar via Wikimedia, CC-BY-SA-4.0 The main challenge is to establish correspondences between ...

Read More »

Benchmarking Robot Manipulation with the Rubik’s Cube

Quantitative evaluations are vital for measuring progress in the field of robot manipulation. Therefore, a recent study on arXiv.org proposes a Rubik’s Cube manipulation benchmark. Rubik’s Cube. Image credit: Max Pixel, CC0 Public Domain In this task, each rotation requires the robot to position its end-effectors with sub-centimeter accuracy. As many such stations are required, errors in the robot’s estimate ...

Read More »

Learning from the brain: Scientist using insights from neuroscience for better artificial intelligence

Machine learning models typically need gigantic data sets and a lot of energy, whereas the brain consumes as much power as a single light bulb. Aalto’s new assistant professor uses neuroscience to make computer programs more efficient. Stéphane Deny started as a new assistant professor in the beginning of December. Image: Mikko Raskinen / Aalto University What do you research ...

Read More »

A Machine Learning Smartphone-based Sensing for Driver Behavior Classification

Road traffic accidents are a growing public health problem. Tracking the driving behavior in real-time and providing feedback on how a driver is behaving on the road is proposed to reduce risky behaviors. Machine learning holds potential of improving driving safety. Image credit: Pxhere, CC0 Public Domain A recent study published on arXiv.org proposes to classify the driving behavior into ...

Read More »

Surprisingly Robust In-Hand Manipulation: An Empirical Study

A recent paper published on arXiv.org proposes a robotic hand with dexterous in-hand manipulation skills. Contact-rich movements like finger-gaiting, pivoting, and the exploitation of gravity are achieved without sensing, hand or object models, or machine learning. Robotic hand used in the study. Image credit: RBO TU Berlin (still image from the YouTube video) A highly compliant hand demonstrates skills which ...

Read More »

Demystifying machine-learning systems

A new method automatically describes, in natural language, what the individual components of a neural network do. Neural networks are sometimes called black boxes because, despite the fact that they can outperform humans on certain tasks, even the researchers who design them often don’t understand how or why they work so well. But if a neural network is used outside the lab, ...

Read More »

Human-Robot Collaborative Carrying of Objects with Unknown Deformation Characteristics

Robotic technologies help to improve productivity and quality in the industry. For instance, robots can collaborate with humans in the transportation of objects. Human-robot collaboration is possible in various tasks, not just industrial. Image credit: Pxhere, CC0 Public Domain This task poses several challenges. Firstly, the robot should share the load with a human effectively. Also, rotations and translations must ...

Read More »

ReSkin: versatile, replaceable, lasting tactile skins

Despite various advancements of AI, it still struggles with dexterous manipulation. Current tactile sensing solutions lack multiple dimensions and fail to scale up. A recent paper on arXiv.org proposes ReSkin – an inexpensive, replaceable, compact, versatile, and long-lasting tactile soft skin. It is composed of soft magnetized skin and a flexible magnetometer-based sensing mechanism. Replacement of the tactile soft skin, ...

Read More »

PointCrack3D: Crack Detection in Unstructured Environments using a 3D-Point-Cloud-Based Deep Neural Network

Cracks on structures are signs of potential weakness and can ultimately lead to catastrophic disasters such as collapses and rockfalls. A recent study on arXiv.org proposes an automated crack detection method that exploits LIDAR data to capture accurate geometric information. Earthquake crack on a highway. Image credit: NPS The researchers propose PointCrack3D, a point-cloud-based DNN approach for crack detection on ...

Read More »