arXiv subjects

Changwen Li

Publications and source records attributed to Changwen Li.

8 recordsLinked to original sources

arXiv2026

Hierarchical Fault Localization for Autonomous Driving Systems with Hypothesis Validation and Intent Analysis

Comprehensive testing is essential for the safety and reliability of Autonomous Driving Systems (ADS). Existing techniques can detect system-level failures or attribute them to coarse-grained modules, but they often fall short of localizing the root cause in source code. As a result, debugging remains labor-intensive, requiring developers to connect behavioral violations with complex implementation logic. To address this gap, we present HINT, a two-phase framework for hierarchical ADS fault localization based on hypothesis validation and intent analysis. In Phase I, HINT transforms failure-triggering execution recordings into multi-modal abstractions and uses causal reasoning to identify the responsible module. In Phase II, it reconstructs design-side intent and implementation-side behavior, then localizes suspicious code through reliability-aware consistency checking, without costly re-simulation. We evaluate HINT on Apollo across diverse failure modes and modules. The results show that HINT achieves the strongest overall performance across module-level diagnosis and code-level localization metrics, with 77.8% end-to-end Class@5 accuracy on real-world bugs.

cs.SE↗

arXiv2026

Validating Generalist Robots with Situation Calculus and STL Falsification

Generalist robots are becoming a reality, capable of interpreting natural language instructions and executing diverse operations. However, their validation remains challenging because each task induces its own operational context and correctness specification, exceeding the assumptions of traditional validation methods. We propose a two-layer validation framework that combines abstract reasoning with concrete system falsification. At the abstract layer, situation calculus models the world and derives weakest preconditions, enabling constraint-aware combinatorial testing to systematically generate diverse, semantically valid world-task configurations with controllable coverage strength. At the concrete layer, these configurations are instantiated for simulation-based falsification with STL monitoring. Experiments on tabletop manipulation tasks show that our framework effectively uncovers failure cases in the NVIDIA GR00T controller, demonstrating its promise for validating general-purpose robot autonomy.

cs.RO↗

arXiv2025

Testing Autonomous Driving Systems -- What Really Matters and What Doesn't

Despite extensive research, the testing of autonomous driving systems (ADS) landscape remains fragmented, and there is currently no basis for an informed technical assessment of the importance and contribution of the current state of the art. This paper attempts to address this problem by exploring two complementary aspects. First, it proposes a framework for comparing existing test methods in terms of their intrinsic effectiveness and validity. It shows that many methods do not meet both of these requirements. Either because they are based on criteria that do not allow for rapid, inexpensive, and comprehensive detection of failures, or because the degree of validity of the properties tested cannot be accurately estimated. In particular, it is shown that most critical test methods do not take into account the nominal operational capabilities of autopilots and generate scenarios that are impossible for the tested vehicles to handle, resulting in unjustified rejections. Secondly, the paper shows that test effectiveness and validity are highly dependent on how autopilots are designed: how they choose between different control policies to perform maneuvers, as well as on the reproducibility of the results. In fact, most test methods take for granted two principles underlying traditional methods, but do not generally apply to ADS. We maintain that the absence of rationality and determinacy significantly impairs the effectiveness and validity of test methods, and provide test results on eight open autopilots, in which most do not satisfy these properties, thereby illustrating this fact. We conclude that under the current state of the art, it is impossible to obtain strong enough guarantees for essential autopilot properties and recommend that autopilots be developed with a view to both rationality and determinacy.

cs.SE↗

arXiv2025

Do Comments and Expertise Still Matter? An Experiment on Programmers' Adoption of AI-Generated JavaScript Code

This paper investigates the factors influencing programmers' adoption of AI-generated JavaScript code recommendations within the context of lightweight, function-level programming tasks. It extends prior research by (1) utilizing objective (as opposed to the typically self-reported) measurements for programmers' adoption of AI-generated code and (2) examining whether AI-generated comments added to code recommendations and development expertise drive AI-generated code adoption. We tested these potential drivers in an online experiment with 173 programmers. Participants were asked to answer some questions to demonstrate their level of development expertise. Then, they were asked to solve a LeetCode problem without AI support. After attempting to solve the problem on their own, they received an AI-generated solution to assist them in refining their solutions. The solutions provided were manipulated to include or exclude AI-generated comments (a between-subjects factor). Programmers' adoption of AI-generated code was gauged by code similarity between AI-generated solutions and participants' submitted solutions, providing a behavioral measurement of code adoption behaviors. Our findings revealed that, within the context of function-level programming tasks, the presence of comments significantly influences programmers' adoption of AI-generated code regardless of the participants' development expertise.

cs.SE↗

arXiv2025

A Comprehensive Evaluation of Four End-to-End AI Autopilots Using CCTest and the Carla Leaderboard

End-to-end AI autopilots for autonomous driving systems have emerged as a promising alternative to traditional modular autopilots, offering the potential to reduce development costs and mitigate defects arising from module composition. However, they suffer from the well-known problems of AI systems such as non-determinism, non-explainability, and anomalies. This naturally raises the question of their evaluation and, in particular, their comparison with existing modular solutions. This work extends a study of the critical configuration testing (CCTest) approach that has been applied to four open modular autopilots. This approach differs from others in that it generates test cases ensuring safe control policies are possible for the tested autopilots. This enables an accurate assessment of the ability to drive safely in critical situations, as any incident observed in the simulation involves the failure of a tested autopilot. The contribution of this paper is twofold. Firstly, we apply the CCTest approach to four end-to-end open autopilots, InterFuser, MILE, Transfuser, and LMDrive, and compare their test results with those of the four modular open autopilots previously tested with the same approach implemented in the Carla simulation environment. This comparison identifies both differences and similarities in the failures of the two autopilot types in critical configurations. Secondly, we compare the evaluations of the four autopilots carried out in the Carla Leaderboard with the CCTest results. This comparison reveals significant discrepancies, reflecting differences in test case generation criteria and risk assessment methods. It underlines the need to work towards the development of objective assessment methods combining qualitative and quantitative criteria.

cs.SE↗

arXiv2024

Rigorous Simulation-based Testing for Autonomous Driving Systems -- Targeting the Achilles' Heel of Four Open Autopilots

Simulation-based testing remains the main approach for validating Autonomous Driving Systems. We propose a rigorous test method based on breaking down scenarios into simple ones, taking into account the fact that autopilots make decisions according to traffic rules whose application depends on local knowledge and context. This leads us to consider the autopilot as a dynamic system receiving three different types of vistas as input, each characterizing a specific driving operation and a corresponding control policy. The test method for the considered vista types generates test cases for critical configurations that place the vehicle under test in critical situations characterized by the transition from cautious behavior to progression in order to clear an obstacle. The test cases thus generated are realistic, i.e., they determine the initial conditions from which safe control policies are possible, based on knowledge of the vehicle's dynamic characteristics. Constraint analysis identifies the most critical test cases, whose success implies the validity of less critical ones. Test coverage can therefore be greatly simplified. Critical test cases reveal major defects in Apollo, Autoware, and the Carla and LGSVL autopilots. Defects include accidents, software failures, and traffic rule violations that would be difficult to detect by random simulation, as the test cases lead to situations characterized by finely-tuned parameters of the vehicles involved, such as their relative position and speed. Our results corroborate real-life observations and confirm that autonomous driving systems still have a long way to go before offering acceptable safety guarantees.

cs.SE↗

arXiv2023

Simulation-based Validation for Autonomous Driving Systems

Simulation is essential to validate autonomous driving systems. However, a simple simulation, even for an extremely high number of simulated miles or hours, is not sufficient. We need well-founded criteria showing that simulation does indeed cover a large fraction of the relevant real-world situations. In addition, the validation must concern not only incidents, but also the detection of any type of potentially dangerous situation, such as traffic violations. We investigate a rigorous simulation and testing-based validation method for autonomous driving systems that integrates an existing industrial simulator and a formally defined testing environment. The environment includes a scenario generator that drives the simulation process and a monitor that checks at runtime the observed behavior of the system against a set of system properties to be validated. The validation method consists in extracting from the simulator a semantic model of the simulated system including a metric graph, which is a mathematical model of the environment in which the vehicles of the system evolve. The monitor can verify properties formalized in a first-order linear temporal logic and provide diagnostics explaining their non satisfaction. Instead of exploring the system behavior randomly as many simulators do, we propose a method to systematically generate sets of scenarios that cover potentially risky situations, especially for different types of junctions where specific traffic rules must be respected. We show that the systematic exploration of risky situations has uncovered many flaws in the real simulator that would have been very difficult to discover by a random exploration process.

cs.SE↗

arXiv2021

ComOpT: Combination and Optimization for Testing Autonomous Driving Systems

ComOpT is an open-source research tool for coverage-driven testing of autonomous driving systems, focusing on planning and control. Starting with (i) a meta-model characterizing discrete conditions to be considered and (ii) constraints specifying the impossibility of certain combinations, ComOpT first generates constraint-feasible abstract scenarios while maximally increasing the coverage of k-way combinatorial testing. Each abstract scenario can be viewed as a conceptual equivalence class, which is then instantiated into multiple concrete scenarios by (1) randomly picking one local map that fulfills the specified geographical condition, and (2) assigning all actors accordingly with parameters within the range. Finally, ComOpT evaluates each concrete scenario against a set of KPIs and performs local scenario variation via spawning a new agent that might lead to a collision at designated points. We use ComOpT to test the Apollo~6 autonomous driving software stack. ComOpT can generate highly diversified scenarios with limited test budgets while uncovering problematic situations such as inabilities to make simple right turns, uncomfortable accelerations, and dangerous driving patterns. ComOpT participated in the 2021 IEEE AI Autonomous Vehicle Testing Challenge and won first place among more than 110 contending teams.

cs.RO↗