AI-based robots and vehicles are expected to operate safely in complex and dynamic environments, even in the presence of component degradation. In such systems, perception relies on sensors such as cameras to capture environmental data, which is then processed by AI models to support decision-making. However, degradation in sensor performance directly impacts input data quality and can impair AI inference. Specifying safety requirements for all possible sensor degradation scenarios leads to unmanageable complexity and inevitable gaps. In this position paper, we present a novel framework that integrates camera noise factor identification with situation coverage analysis to systematically elicit robustness-related safety requirements for AI-based perception systems. We focus specifically on camera degradation in the automotive domain.

Click here to access paper ...

This paper presents a testing approach named SCALOFT for systematically assessing the safety of an autonomous aerial drone in a mine. SCALOFT provides a framework for developing diverse test cases, real-time monitoring of system behaviour, and detection of safety violations. Detected violations are then logged with unique identifiers for detailed analysis and future improvement. SCALOFT helps build a safety argument by monitoring situation coverage and calculating a final coverage measure.

Click here to access paper ...

Underground mines are extremely challenging for autonomous drones as there is limited infrastructure for Simultaneous Localisation and Mapping (SLAM), for the drone to navigate. For example, there is no Global Navigation Satellite System (GNSS), poor lighting, and few distinguishing landmarks. Additionally, the physical environment is extremely harsh, affecting the reliability of the drone. This paper describes the impact of these challenges in designing for, and assuring, safety.

Click here to access paper ...

The three-dimensional swimming tracks of motile microorganisms can be used to identify their species, which holds promise for the rapid identification of bacterial pathogens. Digital holographic microscopy (DHM) is a well-established, but computationally intensive method for obtaining three-dimensional cell tracks from video microscopy data. We accelerate the analysis by an order of magnitude, enabling its use in real time. This technique opens the possibility of rapid identification of bacterial pathogens in drinking water or clinical samples.

Click here to access article ...

Aerial drones are increasingly being considered as a valuable tool for inspection in safety critical contexts. Nowhere is this more true than in mining operations which present a dynamic and dangerous environment for human operators. Drones can be deployed in a number of contexts including efficient surveying as well as search and rescue missions. Operating in these dynamic contexts is challenging however and requires the drones control software to detect and adapt to conditions at run-time.

In this paper we describe a controller framework and simulation environment and provide information on how a user might construct and evaluate their own controllers in the presence of disruptions at run-time.

A virtual machine containing the artifact can be found here: Aloft GitHub repo

The teams will present their system for autonomously and safely recovering a UAV from a mine. Their return-to-home function will activate in the event of a failure (such as remote control loss). Each team will present the return-to-home functionality and an associated safety case. Teams can also demonstrate their software system using a real UAV in the Lab.

Project: Principal Investigator

Unpiloted aerial vehicles (UAVs) are ideal for inspecting infrastructure such as mines. However, the use of UAVs in real-world environments can cause harm to humans, damage the UAVs, and damage infrastructure. Hazards may be caused directly by the UAV if it collides with objects in the environment, or if the UAV fails to complete a mission and needs to be recovered. We need to provide assurance that their use will not cause harm. The three project partners will develop, define, and validate safety requirements and a safe operating concept for multiple UAVs performing mine inspections. This is to ensure safe operation and guarantee early intervention where required.

Click here for project details ...

Springer Nature: Neural Computing and Applications

New article published on using Unity 3D ML Agents and deep reinforcement learning to develop an AI for mapless robot navigation. The article includes a safety assurance assessment of the AI.

Click here to access article ...