Robotics Researcher

Dr. Arman Ibrayeva

Robotics researcher and Principal Investigator with 8 years developing intelligent robotic systems across medical, automotive, nuclear, and agricultural industries. Led end-to-end development from funding acquisition and concept design through hardware prototyping and real-world deployment. 30+ publications in top venues, 5 patents. PhD in Robotic Systems. Deputy Chair of IFToMM Robotics (Kaz).

Currently a Postdoc at Cornell University, Cornell Tech, AI and Robotics Lab, working on human-robot interaction for emergency medicine. Recently developed benchmark models for robot failure detection, and understanding of failure cause from users' facial reactions, modeling longitudinal user adaptation across repeated failures. Now developing new directions on autonomous navigation in safety-critical, crowded environments.

I am searching for new opportunities to grow. Please reach out at ai342@cornell.edu

Walking Robot

Intelligent Walking Robot with Decoupled Motion

Development of deep-learning-based synthesis methods and optimization of an walking robot structural and metric parameters based on functional division of motors. 16 publications including a patent.

Papers View project →

Lower Limb Exoskeleton with Intelligent Biomechanical Feedback Control

Multi-criteria optimized rehabilitation exoskeleton with real-time deep learning-based control utilizing biomechanical signals. 4 journal and conference papers.

Papers Poster View project →

II.

Medical Robots

Causal crowd shaping passage negotiation sequence

Multimodal VLM-Based Crowd Behavior Shaping

Autonomous navigation in dense human crowds by crossing through, not detouring or waiting. The robot treats surrounding people as optimizable variables, not independent obstacles, communicates intent through multimodal cues, and applies minimal nudges that open a passage. A vision-language model supplies open-vocabulary semantic context (surrounding hazards, attention, comfort) on a slow loop; a causal action-outcome predictor and receding-horizon controller execute in real time.

Proposal

HRI Failure Detection in Emergency Medicine

Benchmark models for robot failure detection and categorization from users' facial reactions in emergency department settings. History-aware architecture modeling longitudinal user adaptation. 2 papers under review.

Paper View project →

Pandemic Medical Robotics Complex ($784K)

Robotic complex consisting of a saliva sampling parallel manipulator, patient care assistant, high-payload transporter, and autonomous disinfector. Patented technology.

View project →

Autonomous Robotic Femoral Fracture Reduction

R.ALFRED: parallel-structure robot utilizing 2D X-ray imaging to reconstruct 3D bone models for trajectory planning. Identifies corresponding points on fractured segments and computes optimal surgical trajectories.

View project →

AI-based Sarcopenia and Athletic Injury Analysis

ML/CV musculoskeletal analysis systems for clinical diagnosis from low-quality phone video. Post-injury functional tests and AI-based rehabilitation analysis. 500+ engineered features, 97% accuracy.

View project →

III.

Industrial

Autonomous Agricultural Robot ($200K)

Secured $200K in funding to develop an autonomous field robot with multi-modal perception and deep RL for real-time plant disease detection and localization in greenhouse environments.

View project →

Automotive Robotics for ALLUR ($1M)

Development of five robotic complexes for industrial applications: metal part marking with VIN codes, welding automation (spot, laser, protective gas), and three mobile robotic platforms. Manuscript authored.

View project →

Uranium Production Robotization ($1M/yr)

Robotic complex (RTK-TUK-118) performing the entire cycle of operations from removing and closing the lid, evenly filling HCPU, taking samples, to closing the lid. Replacing manual labor and minimizing radiation exposure.

Report View project →

Novel Rod Pumping Unit for Oil and Gas

Novel design eliminating the massive horse-head balancer, reducing foundation requirements, and improving force transmission. Co-authored 2 patents and 1 journal article.

Paper View project →

IV.

Press & Media

weproject.media

Russian · Interview

On Cornell University, AI research, and myths about poor scientists

Long-form interview on the move to Cornell, the realities of AI and robotics research, and challenging stereotypes about an academic career path.

Read article →

kazpress.kz

English · Feature

When Robots Grow Like Children: A Vision for Safer, Smarter AI

Feature on child-inspired curriculum learning in robotics and what it means to develop AI systems that mature gradually rather than arrive fully formed.

Read article →

kazpress.kz

English · Interview

Medical Robots Must Have Empathy

Conversation on empathy, trust, and the human factors that determine whether robots succeed in clinical and emergency-care environments.

Read article →

kazpress.kz

English · Profile

Building Robots That Heal, Help, and Think Like Us

Profile of cross-disciplinary work spanning rehabilitation exoskeletons, surgical robots, and human-aware navigation in shared environments.

Read article →

Eight outlets

Feb 2025 · Research coverage

Fivefold Cost Reduction for Aerospace-Industry Sensors

Federal science media coverage of collaborative research on piezoelectric thin-film sensors for aviation manufacturing, with new processing methods cutting unit production cost approximately fivefold.

TASS Science Ministry of Science Penza State Univ. CIS Portal AviaPort Innovanews www1.ru Trashbox

Sputnik Kazakhstan

Russian · Research coverage

Russian and Kazakh Scientists Collaborate on Space Research

Coverage of bilateral aerospace and space-research collaboration between scientists in Russia and Kazakhstan, February 2025.

Read article →