Nazim Kemal Ure

Resul Dagdanov, Halil Durmus, Nazim Kemal Ure

In this work, we propose a self-improving artificial intelligence system to enhance the safety performance of reinforcement learning (RL)-based autonomous driving (AD) agents using black-box verification methods. RL algorithms have become popular in AD applications in recent years. However, the performance of existing RL algorithms heavily depends on the diversity of training scenarios. A lack of safety-critical scenarios during the training phase could result in poor generalization performance in real-world driving applications. We propose a novel framework in which the weaknesses of the training set are explored through black-box verification methods. After discovering AD failure scenarios, the RL agent's training is re-initiated via transfer learning to improve the performance of previously unsafe scenarios. Simulation results demonstrate that our approach efficiently discovers safety failures of action decisions in RL-based adaptive cruise control (ACC) applications and significantly reduces the number of vehicle collisions through iterative applications of our method. The source code is publicly available at https://github.com/data-and-decision-lab/self-improving-RL.

Resul Dagdanov, Feyza Eksen, Halil Durmus et al.

Safely navigating through an urban environment without violating any traffic rules is a crucial performance target for reliable autonomous driving. In this paper, we present a Reinforcement Learning (RL) based methodology to DEtect and FIX (DeFIX) failures of an Imitation Learning (IL) agent by extracting infraction spots and re-constructing mini-scenarios on these infraction areas to train an RL agent for fixing the shortcomings of the IL approach. DeFIX is a continuous learning framework, where extraction of failure scenarios and training of RL agents are executed in an infinite loop. After each new policy is trained and added to the library of policies, a policy classifier method effectively decides on which policy to activate at each step during the evaluation. It is demonstrated that even with only one RL agent trained on failure scenario of an IL agent, DeFIX method is either competitive or does outperform state-of-the-art IL and RL based autonomous urban driving benchmarks. We trained and validated our approach on the most challenging map (Town05) of CARLA simulator which involves complex, realistic, and adversarial driving scenarios. The source code is publicly available at https://github.com/data-and-decision-lab/DeFIX

Bengisu Guresti, Abdullah Vanlioglu, Nazim Kemal Ure

Achieving and maintaining cooperation between agents to accomplish a common objective is one of the central goals of Multi-Agent Reinforcement Learning (MARL). Nevertheless in many real-world scenarios, separately trained and specialized agents are deployed into a shared environment, or the environment requires multiple objectives to be achieved by different coexisting parties. These variations among specialties and objectives are likely to cause mixed motives that eventually result in a social dilemma where all the parties are at a loss. In order to resolve this issue, we propose the Incentive Q-Flow (IQ-Flow) algorithm, which modifies the system's reward setup with an incentive regulator agent such that the cooperative policy also corresponds to the self-interested policy for the agents. Unlike the existing methods that learn to incentivize self-interested agents, IQ-Flow does not make any assumptions about agents' policies or learning algorithms, which enables the generalization of the developed framework to a wider array of applications. IQ-Flow performs an offline evaluation of the optimality of the learned policies using the data provided by other agents to determine cooperative and self-interested policies. Next, IQ-Flow uses meta-gradient learning to estimate how policy evaluation changes according to given incentives and modifies the incentive such that the greedy policy for cooperative objective and self-interested objective yield the same actions. We present the operational characteristics of IQ-Flow in Iterated Matrix Games. We demonstrate that IQ-Flow outperforms the state-of-the-art incentive design algorithm in Escape Room and 2-Player Cleanup environments. We further demonstrate that the pretrained IQ-Flow mechanism significantly outperforms the performance of the shared reward setup in the 2-Player Cleanup environment.

Ahmet Semih Tasbas, Safa Onur Sahin, Nazim Kemal Ure

Reinforcement learning (RL) has recently proven itself as a powerful instrument for solving complex problems and even surpassed human performance in several challenging applications. This signifies that RL algorithms can be used in the autonomous air combat problem, which has been studied for many years. The complexity of air combat arises from aggressive close-range maneuvers and agile enemy behaviors. In addition to these complexities, there may be uncertainties in real-life scenarios due to sensor errors, which prevent estimation of the actual position of the enemy. In this case, autonomous aircraft should be successful even in the noisy environments. In this study, we developed an air combat simulation, which provides noisy observations to the agents, therefore, make the air combat problem even more challenging. Thus, we present a state stacking method for noisy RL environments as a noise reduction technique. In our extensive set of experiments, the proposed method significantly outperforms the baseline algorithms in terms of the winning ratio, where the performance improvement is even more pronounced in the high noise levels. In addition, we incorporate a self-play scheme to our training process by periodically updating the enemy with a frozen copy of the training agent. By this way, the training agent performs air combat simulations to an enemy with smarter strategies, which improves the performance and robustness of the agents. In our simulations, we demonstrate that the self-play scheme provides important performance gains compared to the classical RL training.

Umut Demir, Nazim Kemal Ure

In this work we propose a reinforcement learning (RL) framework that controls the density of a large-scale swarm for engaging with adversarial swarm attacks. Although there is a significant amount of existing work in applying artificial intelligence methods to swarm control, analysis of interactions between two adversarial swarms is a rather understudied area. Most of the existing work in this subject develop strategies by making hard assumptions regarding the strategy and dynamics of the adversarial swarm. Our main contribution is the formulation of the swarm to swarm engagement problem as a Markov Decision Process and development of RL algorithms that can compute engagement strategies without the knowledge of strategy/dynamics of the adversarial swarm. Simulation results show that the developed framework can handle a wide array of large-scale engagement scenarios in an efficient manner.

Mehmetcan Kaymaz, Nazim Kemal Ure

Collision avoidance in the presence of dynamic obstacles in unknown environments is one of the most critical challenges for unmanned systems. In this paper, we present a method that identifies obstacles in terms of ellipsoids to estimate linear and angular obstacle velocities. Our proposed method is based on the idea of any object can be approximately expressed by ellipsoids. To achieve this, we propose a method based on variational Bayesian estimation of Gaussian mixture model, the Kyachiyan algorithm, and a refinement algorithm. Our proposed method does not require knowledge of the number of clusters and can operate in real-time, unlike existing optimization-based methods. In addition, we define an ellipsoid-based feature vector to match obstacles given two timely close point frames. Our method can be applied to any environment with static and dynamic obstacles, including the ones with rotating obstacles. We compare our algorithm with other clustering methods and show that when coupled with a trajectory planner, the overall system can efficiently traverse unknown environments in the presence of dynamic obstacles.

Gulay Goktas Sever, Umut Demir, Abdullah Sadik Satir et al.

In this paper, we present a methodology for constructing data-driven maneuver generation models for agile aircraft that can generalize across a wide range of trim conditions and aircraft model parameters. Maneuver generation models play a crucial role in the testing and evaluation of aircraft prototypes, providing insights into the maneuverability and agility of the aircraft. However, constructing the models typically requires extensive amounts of real pilot data, which can be time-consuming and costly to obtain. Moreover, models built with limited data often struggle to generalize beyond the specific flight conditions covered in the original dataset. To address these challenges, we propose a hybrid architecture that leverages a simulation model, referred to as the source model. This open-source agile aircraft simulator shares similar dynamics with the target aircraft and allows us to generate unlimited data for building a proxy maneuver generation model. We then fine-tune this model to the target aircraft using a limited amount of real pilot data. Our approach combines techniques from imitation learning, transfer learning, and reinforcement learning to achieve this objective. To validate our methodology, we utilize real agile pilot data provided by Turkish Aerospace Industries (TAI). By employing the F-16 as the source model, we demonstrate that it is possible to construct a maneuver generation model that generalizes across various trim conditions and aircraft parameters without requiring any additional real pilot data. Our results showcase the effectiveness of our approach in developing robust and adaptable models for agile aircraft.

Umut Demir, A. Sadik Satir, Gulay Goktas Sever et al.

Development of guidance, navigation and control frameworks/algorithms for swarms attracted significant attention in recent years. That being said, algorithms for planning swarm allocations/trajectories for engaging with enemy swarms is largely an understudied problem. Although small-scale scenarios can be addressed with tools from differential game theory, existing approaches fail to scale for large-scale multi-agent pursuit evasion (PE) scenarios. In this work, we propose a reinforcement learning (RL) based framework to decompose to large-scale swarm engagement problems into a number of independent multi-agent pursuit-evasion games. We simulate a variety of multi-agent PE scenarios, where finite time capture is guaranteed under certain conditions. The calculated PE statistics are provided as a reward signal to the high level allocation layer, which uses an RL algorithm to allocate controlled swarm units to eliminate enemy swarm units with maximum efficiency. We verify our approach in large-scale swarm-to-swarm engagement simulations.

Bengisu Guresti, Nazim Kemal Ure

Multi-agent Reinforcement Learning (MARL) problems often require cooperation among agents in order to solve a task. Centralization and decentralization are two approaches used for cooperation in MARL. While fully decentralized methods are prone to converge to suboptimal solutions due to partial observability and nonstationarity, the methods involving centralization suffer from scalability limitations and lazy agent problem. Centralized training decentralized execution paradigm brings out the best of these two approaches; however, centralized training still has an upper limit of scalability not only for acquired coordination performance but also for model size and training time. In this work, we adopt the centralized training with decentralized execution paradigm and investigate the generalization and transfer capacity of the trained models across variable number of agents. This capacity is assessed by training variable number of agents in a specific MARL problem and then performing greedy evaluations with variable number of agents for each training configuration. Thus, we analyze the evaluation performance for each combination of agent count for training versus evaluation. We perform experimental evaluations on predator prey and traffic junction environments and demonstrate that it is possible to obtain similar or higher evaluation performance by training with less agents. We conclude that optimal number of agents to perform training may differ from the target number of agents and argue that transfer across large number of agents can be a more efficient solution to scaling up than directly increasing number of agents during training.

Anil Ozturk, Mustafa Burak Gunel, Resul Dagdanov et al.

Applications of reinforcement learning (RL) are popular in autonomous driving tasks. That being said, tuning the performance of an RL agent and guaranteeing the generalization performance across variety of different driving scenarios is still largely an open problem. In particular, getting good performance on complex road and weather conditions require exhaustive tuning and computation time. Curriculum RL, which focuses on solving simpler automation tasks in order to transfer knowledge to complex tasks, is attracting attention in RL community. The main contribution of this paper is a systematic study for investigating the value of curriculum reinforcement learning in autonomous driving applications. For this purpose, we setup several different driving scenarios in a realistic driving simulator, with varying road complexity and weather conditions. Next, we train and evaluate performance of RL agents on different sequences of task combinations and curricula. Results show that curriculum RL can yield significant gains in complex driving tasks, both in terms of driving performance and sample complexity. Results also demonstrate that different curricula might enable different benefits, which hints future research directions for automated curriculum training.

Reyhan Kevser Keser, Aydin Ayanzadeh, Omid Abdollahi Aghdam et al.

One of the most efficient methods for model compression is hint distillation, where the student model is injected with information (hints) from several different layers of the teacher model. Although the selection of hint points can drastically alter the compression performance, conventional distillation approaches overlook this fact and use the same hint points as in the early studies. Therefore, we propose a clustering based hint selection methodology, where the layers of teacher model are clustered with respect to several metrics and the cluster centers are used as the hint points. Our method is applicable for any student network, once it is applied on a chosen teacher network. The proposed approach is validated in CIFAR-100 and ImageNet datasets, using various teacher-student pairs and numerous hint distillation methods. Our results show that hint points selected by our algorithm results in superior compression performance compared to state-of-the-art knowledge distillation algorithms on the same student models and datasets.

Yunus Bicer, Ali Alizadeh, Nazim Kemal Ure et al.

The objective of this paper is to develop a sample efficient end-to-end deep learning method for self-driving cars, where we attempt to increase the value of the information extracted from samples, through careful analysis obtained from each call to expert driverś policy. End-to-end imitation learning is a popular method for computing self-driving car policies. The standard approach relies on collecting pairs of inputs (camera images) and outputs (steering angle, etc.) from an expert policy and fitting a deep neural network to this data to learn the driving policy. Although this approach had some successful demonstrations in the past, learning a good policy might require a lot of samples from the expert driver, which might be resource-consuming. In this work, we develop a novel framework based on the Safe Dateset Aggregation (safe DAgger) approach, where the current learned policy is automatically segmented into different trajectory classes, and the algorithm identifies trajectory segments or classes with the weak performance at each step. Once the trajectory segments with weak performance identified, the sampling algorithm focuses on calling the expert policy only on these segments, which improves the convergence rate. The presented simulation results show that the proposed approach can yield significantly better performance compared to the standard Safe DAgger algorithm while using the same amount of samples from the expert.

Anil Ozturk, Mustafa Burak Gunel, Melih Dal et al.

Automated lane changing is a critical feature for advanced autonomous driving systems. In recent years, reinforcement learning (RL) algorithms trained on traffic simulators yielded successful results in computing lane changing policies that strike a balance between safety, agility and compensating for traffic uncertainty. However, many RL algorithms exhibit simulator bias and policies trained on simple simulators do not generalize well to realistic traffic scenarios. In this work, we develop a data driven traffic simulator by training a generative adverserial network (GAN) on real life trajectory data. The simulator generates randomized trajectories that resembles real life traffic interactions between vehicles, which enables training the RL agent on much richer and realistic scenarios. We demonstrate through simulations that RL agents that are trained on GAN-based traffic simulator has stronger generalization capabilities compared to RL agents trained on simple rule-driven simulators.

Ali Alizadeh, Majid Moghadam, Yunus Bicer et al.

Autonomous lane changing is a critical feature for advanced autonomous driving systems, that involves several challenges such as uncertainty in other driver's behaviors and the trade-off between safety and agility. In this work, we develop a novel simulation environment that emulates these challenges and train a deep reinforcement learning agent that yields consistent performance in a variety of dynamic and uncertain traffic scenarios. Results show that the proposed data-driven approach performs significantly better in noisy environments compared to methods that rely solely on heuristics.