arXiv subjects

Konstantinos Koufos

Publications and source records attributed to Konstantinos Koufos.

At least 19 recordsLinked to original sources

arXiv2026

Collaborative Multi-Agent Testing for Emergent Failure Discovery in Autonomous Driving Systems

Autonomous Driving Systems (ADS) can fail because of faults within individual modules as well as from interactions across perception, planning, and control. Yet existing ADS testing research often treats key testing functions, such as perturbation generation, behavioural assessment, and test case selection and exploration, as loosely coupled steps rather than coordinated roles for discovering such failures. We present CREAD, a collaborative multi-agent testing framework for testing ADS that organises perturbation generation, behavioural validation, and search coordination through a shared blackboard and an orchestrator. In the current work-in-progress instantiation, the framework focuses on perception-oriented perturbation generation, while remaining extensible to other ADS modules, including planning and control. It currently comprises a Perception Fuzzer Agent, a Metamorphic Validator Agent, and an Orchestrator Agent. Respectively, they generate perturbations, assess behavioural consistency across related scenario pairs, and coordinate further exploration. Experiments in HighwayEnv simulator show that the collaborative configuration improves failure discovery in the highway environment and remains competitive in the roundabout setting. Across the two environments, it yields about 2.1x as many failures per 100 scenarios as the single-agent baseline on average, while gains over a non-collaborative two-agent baseline vary across environments. These results suggest that collaborative multi-agent testing is a promising research direction for emergent ADS behaviour discovery.

cs.SE↗

arXiv2024

Good Data Is All Imitation Learning Needs

In this paper, we address the limitations of traditional teacher-student models, imitation learning, and behaviour cloning in the context of Autonomous/Automated Driving Systems (ADS), where these methods often struggle with incomplete coverage of real-world scenarios. To enhance the robustness of such models, we introduce the use of Counterfactual Explanations (CFEs) as a novel data augmentation technique for end-to-end ADS. CFEs, by generating training samples near decision boundaries through minimal input modifications, lead to a more comprehensive representation of expert driver strategies, particularly in safety-critical scenarios. This approach can therefore help improve the model's ability to handle rare and challenging driving events, such as anticipating darting out pedestrians, ultimately leading to safer and more trustworthy decision-making for ADS. Our experiments in the CARLA simulator demonstrate that CF-Driver outperforms the current state-of-the-art method, achieving a higher driving score and lower infraction rates. Specifically, CF-Driver attains a driving score of 84.2, surpassing the previous best model by 15.02 percentage points. These results highlight the effectiveness of incorporating CFEs in training end-to-end ADS. To foster further research, the CF-Driver code is made publicly available.

cs.CV↗

arXiv2024

Towards A General-Purpose Motion Planning for Autonomous Vehicles Using Fluid Dynamics

General-purpose motion planners for automated/autonomous vehicles promise to handle the task of motion planning (including tactical decision-making and trajectory generation) for various automated driving functions (ADF) in a diverse range of operational design domains (ODDs). The challenges of designing a general-purpose motion planner arise from several factors: a) A plethora of scenarios with different semantic information in each driving scene should be addressed, b) a strong coupling between long-term decision-making and short-term trajectory generation shall be taken into account, c) the nonholonomic constraints of the vehicle dynamics must be considered, and d) the motion planner must be computationally efficient to run in real-time. The existing methods in the literature are either limited to specific scenarios (logic-based) or are data-driven (learning-based) and therefore lack explainability, which is important for safety-critical automated driving systems (ADS). This paper proposes a novel general-purpose motion planning solution for ADS inspired by the theory of fluid mechanics. A computationally efficient technique, i.e., the lattice Boltzmann method, is then adopted to generate a spatiotemporal vector field, which in accordance with the nonholonomic dynamic model of the Ego vehicle is employed to generate feasible candidate trajectories. The trajectory optimising ride quality, efficiency and safety is finally selected to calculate the imminent control signals, i.e., throttle/brake and steering angle. The performance of the proposed approach is evaluated by simulations in highway driving, on-ramp merging, and intersection crossing scenarios, and it is found to outperform traditional motion planning solutions based on model predictive control (MPC).

cs.RO↗

arXiv2024

A Survey on Hybrid Motion Planning Methods for Automated Driving Systems

Motion planning is an essential element of the modular architecture of autonomous vehicles, serving as a bridge between upstream perception modules and downstream low-level control signals. Traditional motion planners were initially designed for specific Automated Driving Functions (ADFs), yet the evolving landscape of highly automated driving systems (ADS) requires motion for a wide range of ADFs, including unforeseen ones. This need has motivated the development of the ``hybrid" approach in the literature, seeking to enhance motion planning performance by combining diverse techniques, such as data-driven (learning-based) and logic-driven (analytic) methodologies. Recent research endeavours have significantly contributed to the development of more efficient, accurate, and safe hybrid methods for Tactical Decision Making (TDM) and Trajectory Generation (TG), as well as integrating these algorithms into the motion planning module. Owing to the extensive variety and potential of hybrid methods, a timely and comprehensive review of the current literature is undertaken in this survey article. We classify the hybrid motion planners based on the types of components they incorporate, such as combinations of sampling-based with optimization-based/learning-based motion planners. The comparison of different classes is conducted by evaluating the addressed challenges and limitations, as well as assessing whether they focus on TG and/or TDM. We hope this approach will enable the researchers in this field to gain in-depth insights into the identification of current trends in hybrid motion planning and shed light on promising areas for future research.

cs.RO↗

arXiv2024

Integrity Monitoring of 3D Object Detection in Automated Driving Systems using Raw Activation Patterns and Spatial Filtering

The deep neural network (DNN) models are widely used for object detection in automated driving systems (ADS). Yet, such models are prone to errors which can have serious safety implications. Introspection and self-assessment models that aim to detect such errors are therefore of paramount importance for the safe deployment of ADS. Current research on this topic has focused on techniques to monitor the integrity of the perception mechanism in ADS. Existing introspection models in the literature, however, largely concentrate on detecting perception errors by assigning equal importance to all parts of the input data frame to the perception module. This generic approach overlooks the varying safety significance of different objects within a scene, which obscures the recognition of safety-critical errors, posing challenges in assessing the reliability of perception in specific, crucial instances. Motivated by this shortcoming of state of the art, this paper proposes a novel method integrating raw activation patterns of the underlying DNNs, employed by the perception module, analysis with spatial filtering techniques. This novel approach enhances the accuracy of runtime introspection of the DNN-based 3D object detections by selectively focusing on an area of interest in the data, thereby contributing to the safety and efficacy of ADS perception self-assessment processes.

cs.CV↗

arXiv2024

SAFE-RL: Saliency-Aware Counterfactual Explainer for Deep Reinforcement Learning Policies

While Deep Reinforcement Learning (DRL) has emerged as a promising solution for intricate control tasks, the lack of explainability of the learned policies impedes its uptake in safety-critical applications, such as automated driving systems (ADS). Counterfactual (CF) explanations have recently gained prominence for their ability to interpret black-box Deep Learning (DL) models. CF examples are associated with minimal changes in the input, resulting in a complementary output by the DL model. Finding such alternations, particularly for high-dimensional visual inputs, poses significant challenges. Besides, the temporal dependency introduced by the reliance of the DRL agent action on a history of past state observations further complicates the generation of CF examples. To address these challenges, we propose using a saliency map to identify the most influential input pixels across the sequence of past observed states by the agent. Then, we feed this map to a deep generative model, enabling the generation of plausible CFs with constrained modifications centred on the salient regions. We evaluate the effectiveness of our framework in diverse domains, including ADS, Atari Pong, Pacman and space-invaders games, using traditional performance metrics such as validity, proximity and sparsity. Experimental results demonstrate that this framework generates more informative and plausible CFs than the state-of-the-art for a wide range of environments and DRL agents. In order to foster research in this area, we have made our datasets and codes publicly available at https://github.com/Amir-Samadi/SAFE-RL.

cs.LG↗

arXiv2024

Run-time Monitoring of 3D Object Detection in Automated Driving Systems Using Early Layer Neural Activation Patterns

Monitoring the integrity of object detection for errors within the perception module of automated driving systems (ADS) is paramount for ensuring safety. Despite recent advancements in deep neural network (DNN)-based object detectors, their susceptibility to detection errors, particularly in the less-explored realm of 3D object detection, remains a significant concern. State-of-the-art integrity monitoring (also known as introspection) mechanisms in 2D object detection mainly utilise the activation patterns in the final layer of the DNN-based detector's backbone. However, that may not sufficiently address the complexities and sparsity of data in 3D object detection. To this end, we conduct, in this article, an extensive investigation into the effects of activation patterns extracted from various layers of the backbone network for introspecting the operation of 3D object detectors. Through a comparative analysis using Kitti and NuScenes datasets with PointPillars and CenterPoint detectors, we demonstrate that using earlier layers' activation patterns enhances the error detection performance of the integrity monitoring system, yet increases computational complexity. To address the real-time operation requirements in ADS, we also introduce a novel introspection method that combines activation patterns from multiple layers of the detector's backbone and report its performance.

cs.CV↗

arXiv2024

Run-time Introspection of 2D Object Detection in Automated Driving Systems Using Learning Representations

Reliable detection of various objects and road users in the surrounding environment is crucial for the safe operation of automated driving systems (ADS). Despite recent progresses in developing highly accurate object detectors based on Deep Neural Networks (DNNs), they still remain prone to detection errors, which can lead to fatal consequences in safety-critical applications such as ADS. An effective remedy to this problem is to equip the system with run-time monitoring, named as introspection in the context of autonomous systems. Motivated by this, we introduce a novel introspection solution, which operates at the frame level for DNN-based 2D object detection and leverages neural network activation patterns. The proposed approach pre-processes the neural activation patterns of the object detector's backbone using several different modes. To provide extensive comparative analysis and fair comparison, we also adapt and implement several state-of-the-art (SOTA) introspection mechanisms for error detection in 2D object detection, using one-stage and two-stage object detectors evaluated on KITTI and BDD datasets. We compare the performance of the proposed solution in terms of error detection, adaptability to dataset shift, and, computational and memory resource requirements. Our performance evaluation shows that the proposed introspection solution outperforms SOTA methods, achieving an absolute reduction in the missed error ratio of 9% to 17% in the BDD dataset.

cs.CV↗

arXiv2023

Counterfactual Explainer Framework for Deep Reinforcement Learning Models Using Policy Distillation

Deep Reinforcement Learning (DRL) has demonstrated promising capability in solving complex control problems. However, DRL applications in safety-critical systems are hindered by the inherent lack of robust verification techniques to assure their performance in such applications. One of the key requirements of the verification process is the development of effective techniques to explain the system functionality, i.e., why the system produces specific results in given circumstances. Recently, interpretation methods based on the Counterfactual (CF) explanation approach have been proposed to address the problem of explanation in DRLs. This paper proposes a novel CF explanation framework to explain the decisions made by a black-box DRL. To evaluate the efficacy of the proposed explanation framework, we carried out several experiments in the domains of automated driving systems and Atari Pong game. Our analysis demonstrates that the proposed framework generates plausible and meaningful explanations for various decisions made by deep underlying DRLs. Source codes are available at: \url{https://github.com/Amir-Samadi/Counterfactual-Explanation}

cs.LG↗

arXiv2023

A Novel Deep Neural Network for Trajectory Prediction in Automated Vehicles Using Velocity Vector Field

Anticipating the motion of other road users is crucial for automated driving systems (ADS), as it enables safe and informed downstream decision-making and motion planning. Unfortunately, contemporary learning-based approaches for motion prediction exhibit significant performance degradation as the prediction horizon increases or the observation window decreases. This paper proposes a novel technique for trajectory prediction that combines a data-driven learning-based method with a velocity vector field (VVF) generated from a nature-inspired concept, i.e., fluid flow dynamics. In this work, the vector field is incorporated as an additional input to a convolutional-recurrent deep neural network to help predict the most likely future trajectories given a sequence of bird's eye view scene representations. The performance of the proposed model is compared with state-of-the-art methods on the HighD dataset demonstrating that the VVF inclusion improves the prediction accuracy for both short and long-term (5~sec) time horizons. It is also shown that the accuracy remains consistent with decreasing observation windows which alleviates the requirement of a long history of past observations for accurate trajectory prediction. Source codes are available at: https://github.com/Amir-Samadi/VVF-TP.

cs.LG↗

arXiv2023

SAFE: Saliency-Aware Counterfactual Explanations for DNN-based Automated Driving Systems

A CF explainer identifies the minimum modifications in the input that would alter the model's output to its complement. In other words, a CF explainer computes the minimum modifications required to cross the model's decision boundary. Current deep generative CF models often work with user-selected features rather than focusing on the discriminative features of the black-box model. Consequently, such CF examples may not necessarily lie near the decision boundary, thereby contradicting the definition of CFs. To address this issue, we propose in this paper a novel approach that leverages saliency maps to generate more informative CF explanations. Source codes are available at: https://github.com/Amir-Samadi//Saliency_Aware_CF.

cs.LG↗

arXiv2023

Multimodal Manoeuvre and Trajectory Prediction for Automated Driving on Highways Using Transformer Networks

Predicting the behaviour (i.e., manoeuvre/trajectory) of other road users, including vehicles, is critical for the safe and efficient operation of autonomous vehicles (AVs), a.k.a., automated driving systems (ADSs). Due to the uncertain future behaviour of vehicles, multiple future behaviour modes are often plausible for a vehicle in a given driving scene. Therefore, multimodal prediction can provide richer information than single-mode prediction, enabling AVs to perform a better risk assessment. To this end, we propose a novel multimodal prediction framework that can predict multiple plausible behaviour modes and their likelihoods. The proposed framework includes a bespoke problem formulation for manoeuvre prediction, a novel transformer-based prediction model, and a tailored training method for multimodal manoeuvre and trajectory prediction. The performance of the framework is evaluated using three public highway driving datasets, namely NGSIM, highD, and exiD. The results show that our framework outperforms the state-of-the-art multimodal methods in terms of prediction error and is capable of predicting plausible manoeuvre and trajectory modes.

cs.LG↗

arXiv2023

Trajectory Prediction with Observations of Variable-Length for Motion Planning in Highway Merging scenarios

Accurate trajectory prediction of nearby vehicles is crucial for the safe motion planning of automated vehicles in dynamic driving scenarios such as highway merging. Existing methods cannot initiate prediction for a vehicle unless observed for a fixed duration of two or more seconds. This prevents a fast reaction by the ego vehicle to vehicles that enter its perception range, thus creating safety concerns. Therefore, this paper proposes a novel transformer-based trajectory prediction approach, specifically trained to handle any observation length larger than one frame. We perform a comprehensive evaluation of the proposed method using two large-scale highway trajectory datasets, namely the highD and exiD. In addition, we study the impact of the proposed prediction approach on motion planning and control tasks using extensive merging scenarios from the exiD dataset. To the best of our knowledge, this marks the first instance where such a large-scale highway merging dataset has been employed for this purpose. The results demonstrate that the prediction model achieves state-of-the-art performance on highD dataset and maintains lower prediction error w.r.t. the constant velocity across all observation lengths in exiD. Moreover, it significantly enhances safety, comfort, and efficiency in dense traffic scenarios, as compared to the constant velocity model.

cs.RO↗

arXiv2022

Two-Hop Connectivity to the Roadside in a VANET Under the Random Connection Model

In this paper, we compute the expected number of vehicles with at least one two-hop path to a fixed roadside unit (RSU) in a multi-hop, one-dimensional vehicular ad hoc network (VANET) where other cars can act as relays. The pairwise channels experience Rayleigh fading in the random connection model, and so exist, with a probability given by a function of the mutual distance between the cars, or between the cars and the RSU. We derive exact expressions for the expected number of cars with a two-hop connection to the RSU when the car density $ρ$ tends to zero and infinity, and determine its behaviour using an infinite oscillating power series in $ρ$, which is accurate for all regimes of traffic density. We also corroborate those findings with a realistic scenario, using snapshots of actual traffic data. Finally, a normal approximation is discussed for the probability mass function of the number of cars with a two-hop connection to the RSU.

cs.NI↗

arXiv2022

Meta Distribution of SIR in the Internet of Things Modelled as a Euclidean Matching

The Poisson bipolar model considers user-base station pairs distributed at random on a flat domain, similar to matchsticks scattered onto a table. Though this is a simple and tractable setting in which to study dense networks, it doesn't properly characterise the stochastic geometry of user-base station interactions in some dense deployment scenarios, which may involve short and long range links, with some paired very nearby optimally, and others sub-optimally due to local crowding. Since the users will pair one-to-one with base stations, we can consider using the popular bipartite Euclidean matching (BEM) from spatial combinatorics, and study the corresponding (meta) distribution of the signal-to-interference-ratio (SIR). This provides detailed information about the proportion of links in the network meeting a target reliability constraint. We can then observe via comparison the impact of taking into account the variable/correlated short-range distances between the transmitter-receiver pairs on the communication statistics. We illustrate and quantify how the widely-accepted bipolar model fails to capture the network-wide reliability of communication in a typical ultra-dense setting based on a binomial point process. We also show how assuming a Gamma distribution for link distances may be a simple improvement on the bipolar model. Overall, BEMs provide good grounds for understanding more sophisticated pairing features in ultra-dense networks.

cond-mat.dis-nn↗

arXiv2021

On the $k$ Nearest-Neighbor Path Distance from the Typical Intersection in the Manhattan Poisson Line Cox Process

In this paper, we consider a Cox point process driven by the Manhattan Poisson line process. We calculate the exact cumulative distribution function (CDF) of the path distance (L1 norm) between a randomly selected intersection and the $k$-th nearest node of the Cox process. The CDF is expressed as a sum over the integer partition function $p\!\left(k\right)$, which allows us to numerically evaluate the CDF in a simple manner for practical values of $k$. These distance distributions can be used to study the $k$-coverage of broadcast signals transmitted from a \ac{RSU} located at an intersection in intelligent transport systems (ITS). Also, they can be insightful for network dimensioning in vehicle-to-everything (V2X) systems, because they can yield the exact distribution of network load within a cell, provided that the \ac{RSU} is placed at an intersection. Finally, they can find useful applications in other branches of science like spatial databases, emergency response planning, and districting. We corroborate the applicability of our distance distribution model using the map of an urban area.

cs.NI↗

arXiv2020

Performance of a Link in a Field of Vehicular Interferers with Hardcore Headway Distance

The Poisson point process (PPP) is not always a realistic model for the locations of vehicles along a road, because it does not account for the safety distance a driver maintains from the vehicle ahead. In this paper, we model the inter-vehicle distance equal to the sum of a constant hardcore distance and a random distance following the exponential distribution. Unfortunately, the probability generating functional of this point process is unknown. To approximate the Laplace transform of interference at the origin, we devise simple approximations for the variance and skewness of interference, and we select suitable probability functions to model the interference distribution. When the coefficient-of-variation and the skewness of interference distribution are high, the PPP (of equal intensity) approximation of the outage probability becomes loose in the upper tail. Relevant scenarios are associated with urban microcells and highway macrocells with a low density of vehicles. The predictions of PPP deteriorate with a multi-antenna maximum ratio combining receiver and temporal indicators related to the performance of retransmission schemes. Our approximations generate good predictions in all considered cases.

cs.NI↗

arXiv2020

Outage in Motorway Multi-Lane VANETs with Hardcore Headway Distance Using Synthetic Traces

In this paper we analyze synthetic mobility traces generated for three-lane unidirectional motorway traffic to find that the locations of vehicles along a lane are better modeled by a hardcore point process instead of the widely-accepted Poisson point process (PPP). In order to capture the repulsion between successive vehicles while maintaining a level of analytical tractability, we make a simple extension to PPP: We model the inter-vehicle distance along a lane equal to the sum of a constant hardcore distance and an exponentially distributed random variable. We calculate the J-function and the Ripley's K-function for this hardcore point process. We fit its parameters to the available traces, and we illustrate that the higher the average speed along a lane, the more prominent the hardcore component becomes. In addition, we consider a transmitter-receiver link on the same lane, and we generate simple formulae for the moments of interference under reduced Palm measure for that lane, and without conditioning for other lanes. We illustrate that under Rayleigh fading a shifted-gamma approximation for the distribution of interference per lane provides a very good fit to the simulated outage probability using the synthetic traces, while the fit using the PPP is poor.

cs.IT↗