Simon Thompson

Yun Li, Simon Thompson, Yidu Zhang et al.

Diffusion-based motion planners have achieved state-of-the-art results on benchmarks such as nuPlan, yet their evaluation within closed-loop production autonomous driving stacks remains largely unexplored. Existing evaluations abstract away ROS 2 communication latency and real-time scheduling constraints, while monolithic ONNX deployment freezes all solver parameters at export time. We present an open-source modular benchmark that addresses both gaps: using ONNX GraphSurgeon, we decompose a monolithic 18,398 node diffusion planner into three independently executable modules and reimplement the DPM-Solver++ denoising loop in native C++. Integrated as a ROS 2 node within Autoware, the open-source AD stack deployed on real vehicles worldwide, the system enables runtime-configurable solver parameters without model recompilation and per-step observability of the denoising process, breaking the black box of monolithic deployment. Unlike evaluations in standalone simulators such as CARLA, our benchmark operates within a production-grade stack and is validated through AWSIM closed-loop simulation. Through systematic comparison of DPM-Solver++ (first- and second-order) and DDIM across six step-count configurations (N in {3, 5, 7, 10, 15, 20}), we show that encoder caching yields a 3.2x latency reduction, and that second-order solving reduces FDE by 41% at N=3 compared to first-order. The complete codebase will be released as open-source, providing a direct path from simulation benchmarks to real-vehicle deployment.

Yun Li, Yidu Zhang, Simon Thompson et al.

Vision-Language-Action (VLA) models for autonomous driving must integrate diverse textual inputs, including navigation commands, hazard warnings, and traffic state descriptions, yet current systems often present these as disconnected fragments, forcing the model to discover on its own which environmental constraints are relevant to the current maneuver. We introduce Causal Scene Narration (CSN), which restructures VLA text inputs through intent-constraint alignment, quantitative grounding, and structured separation, at inference time with zero GPU cost. We complement CSN with Simplex-based runtime safety supervision and training-time alignment via Plackett-Luce DPO with negative log-likelihood (NLL) regularization. A multi-town closed-loop CARLA evaluation shows that CSN improves Driving Score by +31.1% on original LMDrive and +24.5% on the preference-aligned variant. A controlled ablation reveals that causal structure accounts for 39.1% of this gain, with the remainder attributable to information content alone. A perception noise ablation confirms that CSN's benefit is robust to realistic sensing errors. Semantic safety supervision improves Infraction Score, while reactive Time-To-Collision monitoring degrades performance, demonstrating that intent-aware monitoring is needed for VLA systems.

Michael Free, Andrew Langworthy, Mary Dimitropoulaki et al.

The objective of this work is to train a chatbot capable of solving evolving problems through conversing with a user about a problem the chatbot cannot directly observe. The system consists of a virtual problem (in this case a simple game), a simulated user capable of answering natural language questions that can observe and perform actions on the problem, and a Deep Q-Network (DQN)-based chatbot architecture. The chatbot is trained with the goal of solving the problem through dialogue with the simulated user using reinforcement learning. The contributions of this paper are as follows: a proposed architecture to apply a conversational DQN-based agent to evolving problems, an exploration of training methods such as curriculum learning on model performance and the effect of modified reward functions in the case of increasing environment complexity.

Ali Boyali, Simon Thompson, David Robert Wong

Identifying tire and vehicle parameters is an essential step in designing control and planning algorithms for autonomous vehicles. This paper proposes a new method: Simulation-Based Inference (SBI), a modern interpretation of Approximate Bayesian Computation methods (ABC) for parameter identification. The simulation-based inference is an emerging method in the machine learning literature and has proven to yield accurate results for many parameter sets in complex problems. We demonstrate in this paper that it can handle the identification of highly nonlinear vehicle dynamics parameters and gives accurate estimates of the parameters for the governing equations.

Robin Karlsson, David Robert Wong, Simon Thompson et al.

Interconnected road lanes are a central concept for navigating urban roads. Currently, most autonomous vehicles rely on preconstructed lane maps as designing an algorithmic model is difficult. However, the generation and maintenance of such maps is costly and hinders large-scale adoption of autonomous vehicle technology. This paper presents the first self-supervised learning method to train a model to infer a spatially grounded lane-level road network graph based on a dense segmented representation of the road scene generated from onboard sensors. A formal road lane network model is presented and proves that any structured road scene can be represented by a directed acyclic graph of at most depth three while retaining the notion of intersection regions, and that this is the most compressed representation. The formal model is implemented by a hybrid neural and search-based model, utilizing a novel barrier function loss formulation for robust learning from partial labels. Experiments are conducted for all common road intersection layouts. Results show that the model can generalize to new road layouts, unlike previous approaches, demonstrating its potential for real-world application as a practical learning-based lane-level map generator.

Robin Karlsson, David Robert Wong, Kazunari Kawabata et al.

Robust sensing and perception in adverse weather conditions remain one of the biggest challenges for realizing reliable autonomous vehicle mobility services. Prior work has established that rainfall rate is a useful measure for the adversity of atmospheric weather conditions. This work presents a probabilistic hierarchical Bayesian model that infers rainfall rate from automotive lidar point cloud sequences with high accuracy and reliability. The model is a hierarchical mixture of experts model, or a probabilistic decision tree, with gating and expert nodes consisting of variational logistic and linear regression models. Experimental data used to train and evaluate the model is collected in a large-scale rainfall experiment facility from both stationary and moving vehicle platforms. The results show prediction accuracy comparable to the measurement resolution of a disdrometer, and the soundness and usefulness of the uncertainty estimation. The model achieves RMSE 2.42\,mm/h after filtering out uncertain predictions. The error is comparable to the mean rainfall rate change of 3.5\,mm/h between measurements. Model parameter studies show how predictive performance changes with tree depth, sampling duration, and crop box dimension. A second experiment demonstrates the predictability of higher rainfall above 300\,mm/h using a different lidar sensor, demonstrating sensor independence.

Ali Boyali, Simon Thompson

In this paper, we propose an algorithm for optimal generation of nonholonomic paths for planning parking maneuvers with a kinematic car model. We demonstrate the use of Successive Convexification algorithms (SCvx), which guarantee path feasibility and constraint satisfaction, for parking scenarios. In addition, we formulate obstacle avoidance with state-triggered constraints which enables the use of logical constraints in a continuous formulation of optimization problems. This paper contributes to the optimal nonholonomic path planning literature by demonstrating the use of SCvx and state-triggered constraints which allows the formulation of the parking problem as a single optimisation problem. The resulting algorithm can be used to plan constrained paths with cusp points in narrow parking environments.

Simon Thompson, Huiqing Li, Andreas Schumacher

The occurrence of similar code, or `code clones', can make program code difficult to read, modify and maintain. This paper describes industrial case studies of clone detection and elimination using a refactoring and clone detection tool. We discuss how the studies have informed the design of the tool; more importantly, we use the studies to illustrate the complex set of decisions that have to be taken when performing clone elimination in practice. The case studies were performed in collaboration with engineers from Ericsson AB, and used the refactoring tool Wrangler for Erlang. However, the conclusions we draw are largely language-independent, and set out the pragmatics of clone detection and elimination in real-world projects as well as design principles for clone detection decision-support tools.

Dániel Horpácsi, Judit Kőszegi, Simon Thompson

Tool-assisted refactoring transformations must be trustworthy if programmers are to be confident in applying them on arbitrarily extensive and complex code in order to improve style or efficiency. We propose a simple, high-level but rigorous, notation for defining refactoring transformations in Erlang, and show that this notation provides an extensible, verifiable and executable specification language for refactoring. To demonstrate the applicability of our approach, we show how to define and verify a number of example refactorings in the system.