Bio-inspired robots, better algorithms for self-driving cars, and more

Content of the article

From robots inspired by animals and even amoebas to the best algorithms for self-driving cars and robotic surgery, researchers from the University of California San Diego will present a wide range of papers at IROS 2022which returns in a hybrid format from October 23-27, 2022.

IROS is one of the largest and most impactful robotics research conferences in the world. It provides an international forum for the international robotics research community to explore the frontiers of science and technology in intelligent robots and intelligent machines. The theme of IROS 2022 is “Embedded AI for a symbiotic society.” The in-person portion of the conference will take place in Kyoto, Japan.

“We have only recently returned to in-person conferences and it is very exciting that our Contextual Robotics Institute has 16 papers at the conference with comprehensive coverage of mechanisms, control, planning and interaction,” said Henrik Christensen, Director of CRI and Professor in the Department of Computer Science and Engineering. “We are not only back in person, but also with a stronger set of documents, which is very gratifying.”

Here are the papers that researchers from the departments of Computer Science, Electrical and Computer Engineering, and Mechanical and Aerospace Engineering are contributing this year.

Nikolai Atanasov

Safe control synthesis with uncertain dynamics and constraints
K. Long, V. Dhiman, M. Leok, J. Cortés and N. Atanasov

DARL1N: Distributed Multi-Agent Reinforcement Learning with Single-Hop Neighbors
B. Wang, J. Xie and N. Atanasov

Active mapping via Shannon mutual information gradient ascent optimization on continuous SE(3) trajectories
A. Asgharivaskasi, S. Koga and N. Atanasov,

WFA-IRL: Inverse Reinforcement Learning of Autonomous Behaviors Coded as Weighted Finite Automata
T. Wang and N. Atanasov

Sean Gao

Quantifying the safety of learning-based autonomous driving control using quasi-barrier functions

Authors: Zhizhen Qin, Tsui-Wei Weng, Sicun Gao

Nick Gravish

Amoeba-inspired swim through isoperimetric body shape modulation

C Sparks, N Justus, R Hatton, N Gravish

Conformal thorax design for robustness and elastic energy exchange in flapping-wing robots
Hang Gao, James Lynch and Nick Gravish

Sylvia Herbert

Refine Control Barrier Functions with Hamilton-Jacobi Accessibility

Sander Tonkens and Sylvia Herbert

Tania Morimoto

Training non-parametric models in real time via the generalized expert online product

Connor Watson and Tania K. Morimoto

Design and evaluation of a miniaturized force sensor based on wave backscatter

Daegue Park, Agrim Gupta, Shayaun Bashar, Cédric Girerd, Dinesh Bharadia and Tania. K.Morimoto

Tactile perception for growing robots via discrete curvature measurements

Micah Bryant, Connor Watson, Tania K. Morimoto

Mike Tolley

Active suction locomotion in a soft robot inspired by starfish

Ishida M, Sandoval JA, Lee S, Huen S, Tolley MT (2022)

xialong wang

From one hand to many hands: learning by imitation for dexterous manipulation from single-camera teleoperation.

Yuzhe Qin, Hao Su, Xiaolong Wang

Online adaptation for implicit object tracking and shape reconstruction in nature.

Jianglong Ye, Yuntao Chen, Naiyan Wang, Xiaolong Wang.

Vision-guided quadrupedal locomotion in nature with multimodal delay randomization.

Chieko Sarah Imai, Minghao Zhang, Yuchen Zhang, Marcin Kierebiński, Ruihan Yang, Yuzhe Qin, Xiaolong Wang.

Press release

Michael Yip

6D Markerless Suture Needle Pose Tracking with Robust Uncertainty Estimation for Autonomous Minimally Invasive Robotic Surgery

Zih-Yun Chiu, Albert Z. Lia, Florian Richter, Bjorn Johnson, and Michael C. Yip

Video

Sharon D. Cole