A Local Agentic AI Framework to Generate Edge Case Scenarios for Safety Assurance

Victoria J. Hodge; Colin Paterson; Rob Alexander

cs.AIOct 9, 2026

A Local Agentic AI Framework to Generate Edge Case Scenarios for Safety Assurance

Authors: Victoria J. Hodge, Colin Paterson, Rob Alexander

Organizations: Department of Computer Science, University of York, York, YO10 5DD, UK

Abstract

Generating edge case scenarios is crucial for training, testing and safety assuring autonomous systems. This identifies hazardous failures to increase system reliability and robustness, ensures the system can still function safely on the boundaries of its operational design domain and learns the expected system behaviour for these edge cases. The rapid advancement in the generation and analysis capabilities of language models (LMs) has expanded their applicability and recent work has investigated LM scenario generation. However, many recent approaches rely on human-in-the-loop or rule-based strategies. This restricts the LM's ability to generate diverse and edge case scenarios; the long tail of the distribution. Our work uses local LMs in a laptop-based agentic AI framework for adaptive scenario generation. This is cheap, fully-controlled, ensures data privacy and security. It uses think, act, observe, update, and continue loops with lightweight JSON encoding to generate scenarios. We use the system requirements, system description and environment description as knowledge bases in contextual prompts for an LM to ensure scenario realism. A second, independent LM critiques each scenario using domain specific criteria so it can be refined. The refined scenarios are run in simulation and evaluation metrics collected. These metric scores are analysed and similar scenarios are created for edge cases (exploitation) or novel scenarios are generated to explore the scenario space. Post-hoc clustering using metric scores and thematic-type analysis of the rich-text scenario descriptions produced by the LM pinpoints functional gaps and their triggering conditions. This creates traceable evidence for the safety assurance case. Our work highlights the potential of a flexible and adaptive local agentic AI framework for automating edge case generation and producing traceable safety assurance evidence.

Figures & tables

Figure 1: The overall architecture and process flow of our proposed agentic AI framework. Icons from https://thenounproject.com/ , the Noun Project

Figure 2: Overview of the first part of the agentic AI scenario generation framework. Part 1 produces the JSON-encoded scenarios to run in simulation in part 2 which is shown in Figure 3 . The domain ontology allows us to map the sections of the operational deign domain description to sections of the LM prompt and is described in section 4.2 .

Figure 3: Overview of the agentic AI simulation, evaluation and decision logic to assess the scenarios and to decide whether to exploit an existing scenario or explore novel scenarios by looping back to step 1 in Figure 2 .

Figure 4: An example scenario laid out in the Sorter arena in Unity-3D. The blue cube in the centre is the robotic agent. The red crosses are random obstacles and the coloured cubes are fixed obstacles.

Figure 5: High-level goal structuring notation for the robotic-agent navigating the circular arena. From the claim in G1, we must demonstrate that the risk of colliding with obstacles is as low as reasonably practicable. The GSN decomposes through levels includingt the GSN shown in Figure 6

Figure 6: Lower level goal structuring notation fragment for the robotic-agent’s navigation. This paper focuses on generating the evidence for Sn1.4 - Sn1.7. After generating our edge case scenarios and analysing the evaluation metrics and simulation logs, we can instantiate the artefacts for Sn1.4 { eval metrics }, Sn1.5 { simulation failures & collisions log (

E_{sim}

) }, Sn1.6 { identified SOTIF area 3

\rightarrow

area 2 edge cases } & Sn1.7 { hazard discovery report }

Figure 7: Eight scenarios generated by the Agentic AI and laid out in the Sorter arena in Unity-3D.

Baseline	Description
Random_1	Random asset placement generated using the Python random number generator with 1-20 tiles and 1-10 obstacles. Tile values are generated as unique values (used at most once). If tile or obstacle locations clash then a new location is generated.
Random_5	Random asset placement generated using the Python random number generator with 1-20 tiles and 5 -10 obstacles. Tile values are generated as unique values (used at most once). If tile or obstacle locations clash then a new location is generated.

Table 1: The baseline scenario generation approaches to allow us to evaluate the merits of our agentic approach.

Metric	Description
Success score	how many times it successfully completed the task (over 50 episodes)
Timeout score	how many times the task timed out (over 50 episodes)
Wrong tile score	how many times it visited the wrong tile (over 50 episodes)
Avg path length	the average number of steps taken (over 50 episodes)
Ideal path length	the ideal number of steps (shortest path)
Regret	(avgPathLength / idealPathLength)

Table 2: The set of evaluation metrics exported by Unity-3D for use in the agentic AI closed loop reflection analyses.

Figure 14: Chart showing the average path length across all scenarios for each scenario generation method. The path length is 0 if it visited the wrong tile.

Panel of Figure 17 — Figure 17: Spatial maps of obstacle placements across all scenarios for each scenario generation method.

Panel of Figure 18 — Figure 18: Chart showing how many times each number of obstacles was selected across all scenarios for each scenario generation method.

Panel of Figure 19 — Figure 19: Chart showing how many times each number (count) of tiles was selected across all scenarios for each scenario generation method.

Panel of Figure 20 — Figure 20: Chart showing how many times each tile position was selected across all scenarios for each scenario generation method.

Panel of Figure 21 — Figure 21: Chart showing how many times each tile value was selected across all scenarios for each scenario generation method.

Figure 22: Chart showing the ideal path length across all scenarios for each scenario generation method.

JSON Parameter / Asset Configuration	ISO 21448 Triggering Condition (TC)	Exposed Functional Gaps (FG)	Resulting Unsafe Behaviour
Dense clusters of obstacles near the robot’s starting point	Algorithmic search space saturation: Complex spatial topology	Planning oscillations ( $FG_{plan}$ ): Local planner path deadlock or trajectory freezing.	Timeouts: Robot stuck, path oscillation, or trajectory abandonment
Tile access blockades (clearance $<w_{robot}$ )	Geometric Constraint Boundary: Access to tile blocked requiring robot to reverse	Perception Insufficiency ( $SW_{per}$ ): LiDAR point-cloud sparsity (no backward facing LiDAR)	Collisions with wrong tile or arena walls, or obstacle contact.

Table 3: Hazard Discovery Report: formal mapping table to demonstrate how the generated spatial layouts for

\theta_{JSON_{1}}

and

\theta_{JSON_{2}}

map to triggering conditions and functional gaps exposed by the simulation’s evaluation metrics. We assume that timeouts are a safety risk as the robot is stuck and may oscillate leading to collisions.

AI Assurance

Artificial intelligence assurance develops evidence and controls to evaluate, govern, and certify AI systems across their lifecycle and deployment contexts.

See all 77 papers

EvoDrive: Pareto Evolution for Safety-Critical Autonomous Driving via Self-Improving LLM Agents
Jun 2, 2026Tong Nie, Yuewen Mei, Yihong Tang +4Multi-Objective Evolutionary Optimization Autonomous Driving Safety Evaluation
The Hong Kong Polytechnic University · Tongji University · McGill University +1
PCASim: Promptable Closed-loop Adversarial Simulation for Urban Traffic Environment
May 15, 2026Chuancheng Zhang, Zhenhao Wang, Kaizheng Li +3Autonomous Driving Safety Evaluation Synthetic Data Generation
Shenzhen Research Institute of Shandong University, Shenzhen, 518057, Guangdong, China · School of Airspace Science and Engineering, Shandong University, 264209 Weihai, China. · School of Computer Science and Technology, Shandong University of Finance and Economics, 250014, Jinan, China.
ForesightSafety-SAGE:A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents
Jun 7, 2026Lu Jia, Haibo Tong, Feifei Zhao +5AI Agent Safety Benchmarks LLM Agent Safety
BrainCog AI Lab, Institute of Automation, Chinese Academy of Sciences · Beijing Institute of AI Safety and Governance, China · University of Chinese Academy of Sciences (UCAS), Beijing, China +2

Jun 2, 2026J/K move · Enter open · S save

EvoDrive: Pareto Evolution for Safety-Critical Autonomous Driving via Self-Improving LLM Agents

Tong Nie, Yuewen Mei, Yihong Tang, Junlin He+3

The Hong Kong Polytechnic University · Tongji University · McGill University+1

Multi-Objective Evolutionary Optimization Autonomous Driving Safety Evaluation Adversarial Scenario Generation Autonomous Driving Self-Improving Agents Autonomous Driving Simulation+1

Open paper AskPDF arXiv

Generating safety-critical scenarios is essential for validating and improving autonomous driving systems, yet it inherently requires maximizing adversariality to expose failures while preserving realism. Existing methods usually manage this trade-off with handcrafted heuristics, confining generation to known priors and overlooking underexplored patterns. While recent open-ended agentic evolution can push this limit, unconstrained general agents lack strict simulator grounding and tend to collapse the multi-objective tension into single-scalar maximization. Here we present EvoDrive, the first automated, LLM-based agentic evolution framework for multi-objective scenario generation. EvoDrive employs a simulator-grounded actor-critic architecture where a memory-driven actor iteratively proposes improvements to the generators and critics filter out implausible candidates, and a self-evolving world evaluator routes promising proposals to optimize simulation budgets. EvoDrive further maintains a Pareto archive of evaluated candidates to preserve diverse attack-realism trade-offs and guide future evolution via simulation feedback. Benchmark results on MetaDrive and CARLA show that EvoDrive not only significantly expands the Pareto frontier across various generators, but also produces valuable scenarios for policy training.

Abstract

AI Assurance

Explore similar work

EvoDrive: Pareto Evolution for Safety-Critical Autonomous Driving via Self-Improving LLM Agents

PCASim: Promptable Closed-loop Adversarial Simulation for Urban Traffic Environment

ForesightSafety-SAGE:A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents