Christopher M. Poskitt

About

Christopher M. Poskitt is a leading researcher at the intersection of artificial intelligence and safety-critical cyber-physical systems, with a primary focus on the security and robustness of collaborative robots (cobots). His work addresses the unique vulnerabilities of AI-enabled systems that operate in close proximity to humans, particularly in smart factory environments. Poskitt’s major contributions include pioneering research on physical adversarial attacks against robotic arms, demonstrating how subtle manipulations can deceive object detection models in real-world settings—a study that has garnered 23 citations and highlighted critical safety gaps. He has also advanced defensive techniques, such as boundary data selection for adversarial training, which enhances model resilience without sacrificing the rapid response times essential for human-robot collaboration. This work, published in 2023, is already shaping best practices for securing AI in manufacturing. Poskitt’s research is notable for bridging the gap between theoretical adversarial machine learning and practical, high-stakes deployments, earning him recognition as a key voice in the emerging field of trustworthy AI for robotics. His findings are vital for engineers and policymakers working to ensure that cobots remain both intelligent and safe.

Research Focus

Key Achievements

Top Papers

1. 1
   Physical Adversarial Attack on a Robotic Arm

   23 citations · 2022
2. 2
   Boosting Adversarial Training in Safety-Critical Systems Through Boundary Data Selection

   4 citations · 2023

Key Collaborators

Contact & Links