Team NimbRo's robot Mario won the MBZIRC 2017 Challenge 2 by autonomously manipulating a valve stem using a wrench. The robot uses an omnidirectional base for locomotion and a 3D laser scan detector to find the manipulation panel. A deep neural network detects and selects the correct tool from grayscale images, and motion primitives are adapted to turn the valve stem. Why it matters: This work demonstrates advanced robotic manipulation capabilities relevant for industrial automation and hazardous environment operations in the region.
Lorenzo Jamone from Queen Mary University of London presented on cognitive robotics, focusing on tactile exploration and manipulation by robots. The talk covered combining biology, engineering, and AI for advanced robotic systems. Jamone directs the CRISP group and has over 100 publications in cognitive robotics. Why it matters: This highlights the ongoing research into more sophisticated robotic systems that can interact with complex environments, an area crucial for future applications in manufacturing and human-robot collaboration in the GCC.
A KAUST team led by Hossein Fariborzi won second place in the MEMS Design Contest for their "MEMS Resonator for Oscillator, Tunable Filter and Re-Programmable Logic Applications." The device is runtime-reprogrammable, allowing the function of each device in the circuit to be changed during operation. The KAUST team demonstrated that two MEMS resonators could replace over 20 transistors in applications like digital adders, reducing digital circuit complexity. Why it matters: This innovation could significantly reduce power consumption, chip area, and manufacturing costs in microprocessors, advancing the development of energy-efficient microcomputers in the region.
Patrick van der Smagt, Director of AI Research at Volkswagen Group, discussed the use of generative machine learning models for predicting and controlling complex stochastic systems in robotics. The talk highlighted examples in robotics and beyond and addressed the challenges of achieving quality and trust in AI systems. He also mentioned his involvement in a European industry initiative on trust in AI and his membership in the AI Council of the State of Bavaria. Why it matters: Understanding control in robotics, along with trust in AI, are key issues for further development of autonomous systems, especially in industrial applications within the GCC region.
Prof. Simon Gröblacher from Delft University of Technology presented a seminar on using mechanical systems in quantum information processing, focusing on their potential as quantum memories and transducers. The seminar highlighted experiments demonstrating non-classical behavior of mechanical motion by coupling a micro-fabricated acoustic resonator to single optical photons. Quantum control over acoustic motion was established, including the generation and readout of single phononic excitations, along with light-matter entanglement. Why it matters: This research advances the use of micro-fabricated acoustic resonators for quantum information processing and fundamental tests of quantum physics.
KAUST researchers developed VENTIBAG, a mobile AI-powered ventilator, in response to the COVID-19 pandemic. The device extracts and delivers pure oxygen, adjusting support based on real-time monitoring of the patient's condition via cloud connectivity. Funded by a KAUST Innovation Challenge grant, the portable ventilator is now advancing to the testing stage for medical applications. Why it matters: This innovation addresses critical needs for remote patient care and reducing hospital overcrowding, particularly relevant in resource-constrained environments.
Researchers at MBZUAI introduce "Interactive Video Reasoning," a new paradigm enabling models to actively "think with videos" by performing iterative visual actions to gather and refine evidence. They developed Video CoM, which reasons through a Chain of Manipulations (CoM), and constructed Video CoM Instruct, an 18K instruction tuning dataset for multi-step manipulation reasoning. The model is further optimized via reinforcement learning with reasoning aware Group Relative Policy Optimization (GRPO), achieving strong results across nine video reasoning benchmarks.