It is useful for autonomous vehicles to have the ability to infer the goals of other vehicles (goal recognition), in order to safely interact with other vehicles and predict their future trajectories. Goal recognition methods must be fast to run in real time and make accurate inferences. As autonomous driving is safety-critical, it is important to have methods which are human interpretable and for which safety can be formally verified. Existing goal recognition methods for autonomous vehicles fail to satisfy all four objectives of being fast, accurate, interpretable and verifiable. We propose Goal Recognition with Interpretable Trees (GRIT), a goal recognition system for autonomous vehicles which achieves these objectives. GRIT makes use of decision trees trained on vehicle trajectory data. Evaluation on two vehicle trajectory datasets demonstrates the inference speed and accuracy of GRIT compared to an ablation and two deep learning baselines. We show that the learned trees are human interpretable and demonstrate how properties of GRIT can be formally verified using an SMT solver.
Sharing parameters in multi-agent deep reinforcement learning has played an essential role in allowing algorithms to scale to a large number of agents. Parameter sharing between agents significantly decreases the number of trainable parameters, shortening training times to tractable levels, and has been linked to more efficient learning. However, having all agents share the same parameters can also have a detrimental effect on learning. We demonstrate the impact of parameter sharing methods on training speed and converged returns, establishing that when applied indiscriminately, their effectiveness is highly dependent on the environment. Therefore, we propose a novel method to automatically identify agents which may benefit from sharing parameters by partitioning them based on their abilities and goals. Our approach combines the increased sample efficiency of parameter sharing with the representational capacity of multiple independent networks to reduce training time and increase final returns.
Current methods for authentication based on public-key cryptography are vulnerable to quantum computing. We propose a novel approach to authentication in which communicating parties are viewed as autonomous agents which interact repeatedly using their private decision models. The security of this approach rests upon the difficulty of learning the model parameters of interacting agents, a problem which we conjecture is also hard for quantum computing. We develop methods which enable a server agent to classify a client agent as either legitimate or adversarial based on their past interactions. Moreover, we use reinforcement learning techniques to train server policies which effectively probe the client's decisions to achieve more sample-efficient authentication, while making modelling attacks as difficult as possible via entropy-maximization principles. We empirically validate our methods for authenticating legitimate users while detecting different types of adversarial attacks.
Ad hoc teamwork is the challenging problem of designing an autonomous agent which can adapt quickly to collaborate with previously unknown teammates. Prior work in this area has focused on closed teams in which the number of agents is fixed. In this work, we consider open teams by allowing agents of varying types to enter and leave the team without prior notification. Our proposed solution builds on graph neural networks to learn scalable agent models and value decompositions under varying team sizes, which can be jointly trained with a reinforcement learning agent using discounted returns objectives. We demonstrate empirically that our approach results in agent policies which can robustly adapt to dynamic team composition, and is able to effectively generalize to larger teams than were seen during training.
Modelling the behaviours of other agents (opponents) is essential for understanding how agents interact and making effective decisions. Existing methods for opponent modelling commonly assume knowledge of the local observations and chosen actions of the modelled opponents, which can significantly limit their applicability. We propose a new modelling technique based on variational autoencoders which uses only the local observations of the agent under control: its observed world state, chosen actions, and received rewards. The model is jointly trained with the agent's decision policy using deep reinforcement learning techniques. We provide a comprehensive evaluation and ablation study in diverse multi-agent tasks, showing that our method achieves significantly higher returns than a baseline method which does not use opponent modelling, and comparable performance to an ideal baseline which has full access to opponent information.
Multi-agent deep reinforcement learning (MARL) suffers from a lack of commonly-used evaluation tasks and criteria, making comparisons between approaches difficult. In this work, we evaluate and compare three different classes of MARL algorithms (independent learners, centralised training with decentralised execution, and value decomposition) in a diverse range of multi-agent learning tasks. Our results show that (1) algorithm performance depends strongly on environment properties and no algorithm learns efficiently across all learning tasks; (2) independent learners often achieve equal or better performance than more complex algorithms; (3) tested algorithms struggle to solve multi-agent tasks with sparse rewards. We report detailed empirical data, including a reliability analysis, and provide insights into the limitations of the tested algorithms.
Exploration in multi-agent reinforcement learning is a challenging problem, especially in environments with sparse rewards. We propose a general method for efficient exploration by sharing experience amongst agents. Our proposed algorithm, called Shared Experience Actor-Critic (SEAC), applies experience sharing in an actor-critic framework. We evaluate SEAC in a collection of sparse-reward multi-agent environments and find that it consistently outperforms two baselines and two state-of-the-art algorithms by learning in fewer steps and converging to higher returns. In some harder environments, experience sharing makes the difference between learning to solve the task and not learning at all.
The ability to predict the intentions and driving trajectories of other vehicles is a key problem for autonomous driving. We propose an integrated planning and prediction system which leverages the computational benefit of using a finite space of maneuvers, and extend the approach to planning and prediction of sequences (plans) of maneuvers via rational inverse planning to recognise the goals of other vehicles. Goal recognition informs a Monte Carlo Tree Search (MCTS) algorithm to plan optimal maneuvers for the ego vehicle. Inverse planning and MCTS utilise a shared set of defined maneuvers to construct plans which are explainable by means of rationality, i.e. plans are optimal in given metrics. Evaluation in simulations of four urban driving scenarios demonstrate the system's ability to robustly recognise the goals of other vehicles while generating near-optimal plans. In each scenario we extract intuitive explanations for the recognised goals and maneuver predictions which justify the system's decisions.
Lessons learned from the increasing diversity of road trial deployments of autonomous vehicles have made clear that guaranteeing safety of driving decisions is a crucial bottleneck on the path towards wider adoption. A promising direction is to pose safety requirements as planning constraints in nonlinear optimization problems for motion synthesis. However, many implementations of this approach are limited by uncertain convergence and local optimality of the solutions achieved, affecting overall robustness. In this paper, we propose a novel two-stage optimization framework: in the first stage, we find a global but approximate solution to a Mixed-Integer Linear Programming (MILP) formulation of the motion synthesis problem, the output of which initializes a second Nonlinear Programming (NLP) stage. The MILP stage enforces hard constraints including safety and road rules, while the NLP stage refines that solution within safety bounds to make it feasible with respect to vehicle dynamics and smoothness. We demonstrate the usefulness of our framework through experiments in complex driving situations, showing it outperforms a state of the art baseline in terms of convergence, comfort and progress metrics.
Multi-agent systems exhibit complex behaviors that emanate from the interactions of multiple agents in a shared environment. In this work, we are interested in controlling one agent in a multi-agent system and successfully learn to interact with the other agents that have fixed policies. Modeling the behavior of other agents (opponents) is essential in understanding the interactions of the agents in the system. By taking advantage of recent advances in unsupervised learning, we propose modeling opponents using variational autoencoders. Additionally, many existing methods in the literature assume that the opponent models have access to opponent's observations and actions during both training and execution. To eliminate this assumption, we propose a modification that attempts to identify the underlying opponent model using only local information of our agent, such as its observations, actions, and rewards. The experiments indicate that our opponent modeling methods achieve equal or greater episodic returns in reinforcement learning tasks against another modeling method.