Oxford Dynamics Strider: the robot navigating hazardous terrain
Oxford Dynamics, a UK-based AI startup, has been crafting a robot designed to undertake tasks in hazardous conditions which have been traditionally performed by humans.
Oxford Dynamics, underpinned by a £1 million contract from the UK Ministry of Defence (MoD), has developed the robot, Strider.
The project aims to enhance safety in environments plagued by chemical, biological, or nuclear threats, and areas with lethal radiation exposure.
Strider possesses the capability to collect contaminated materials, secure them in containers, and execute semi-autonomous tasks. The robot is able to navigate difficult terrains utilising cutting-edge technologies such as infrared, radar, and lidar (light detection and ranging).
Oxford Dynamics explained in a post on LinkedIn: “The tracked robot, under development at Harwell Science and Innovation Campus, is intended to replace people from being the first point of contact in the event of hazardous incidents and is being developed for end-customer the Department for Environment, Food and Rural Affairs. STRIDER is ultimately intended to assist Defra in their role encompassing environmental remediation and recovery following Chemical, Biological, Radiological and Nuclear (CBRN) incidents in the UK.”
The company plans to enhance Strider’s capabilities by integrating its AVIS AI software, which stands for A Very Intelligent System, inspired by JARVIS from the Iron Man films.
AVIS represents an advanced AI-powered automated visual inspection solution that transforms the inspection process. It boosts inspection volume, minimises missed defects, and significantly reduces operating and capital expenses.
Unlike conventional AI development methods, AVIS removes the need for highly skilled, expensive, and scarce talent resources.
Oxford Dynamics is also developing several future AI tools designed to handle data in a manner that mimics human cognition. This approach alleviates cognitive load and maintains the unique value of complex data, which is often lost with traditional data processing methods, particularly when dealing with visual data.