Tackling Real-World Autonomous Driving using Deep Reinforcement Learning

In the typical autonomous driving stack, planning and control systems represent two of the most crucial components in which data retrieved by sensors and processed by perception algorithms are used to implement a safe and comfortable self-driving behavior. In particular, the planning module predicts the path the autonomous car should follow taking the correct high-level maneuver, while control systems perform a sequence of low-level actions, controlling steering angle, throttle and brake. In this work, we propose a model-free Deep Reinforcement Learning Planner training a neural network that predicts both acceleration and steering angle, thus obtaining a single module able to drive the vehicle using the data processed by localization and perception algorithms on board of the self-driving car. In particular, the system that was fully trained in simulation is able to drive smoothly and safely in obstacle-free environments both in simulation and in a real-world urban area of the city of Parma, proving that the system features good generalization capabilities also driving in those parts outside the training scenarios. Moreover, in order to deploy the system on board of the real self-driving car and to reduce the gap between simulated and real-world performances, we also develop a module represented by a tiny neural network able to reproduce the real vehicle dynamic behavior during the training in simulation.

End-to-End Intersection Handling using Multi-Agent Deep Reinforcement Learning

Navigating through intersections is one of the main challenging tasks for an autonomous vehicle. However, for the majority of intersections regulated by traffic lights, the problem could be solved by a simple rule-based method in which the autonomous vehicle behavior is closely related to the traffic light states. In this work, we focus on the implementation of a system able to navigate through intersections where only traffic signs are provided. We propose a multi-agent system using a continuous, model-free Deep Reinforcement Learning algorithm used to train a neural network for predicting both the acceleration and the steering angle at each time step. We demonstrate that agents learn both the basic rules needed to handle intersections by understanding the priorities of other learners inside the environment, and to drive safely along their paths. Moreover, a comparison between our system and a rule-based method proves that our model achieves better results especially with dense traffic conditions. Finally, we test our system on real world scenarios using real recorded traffic data, proving that our module is able to generalize both to unseen environments and to different traffic conditions.

From Simulation to Real World Maneuver Execution using Deep Reinforcement Learning

Deep Reinforcement Learning has proved to be able to solve many control tasks in different fields, but the behavior of these systems is not always as expected when deployed in real-world scenarios. This is mainly due to the lack of domain adaptation between simulated and real-world data together with the absence of distinction between train and test datasets. In this work, we investigate these problems in the autonomous driving field, especially for a maneuver planning module for roundabout insertions. In particular, we present a system based on multiple environments in which agents are trained simultaneously, evaluating the behavior of the model in different scenarios. Finally, we analyze techniques aimed at reducing the gap between simulated and real-world data showing that this increased the generalization capabilities of the system both on unseen and real-world scenarios.

Intelligent Roundabout Insertion using Deep Reinforcement Learning

An important topic in the autonomous driving research is the development of maneuver planning systems. Vehicles have to interact and negotiate with each other so that optimal choices, in terms of time and safety, are taken. For this purpose, we present a maneuver planning module able to negotiate the entering in busy roundabouts. The proposed module is based on a neural network trained to predict when and how entering the roundabout throughout the whole duration of the maneuver. Our model is trained with a novel implementation of A3C, which we will call Delayed A3C (D-A3C), in a synthetic environment where vehicles move in a realistic manner with interaction capabilities. In addition, the system is trained such that agents feature a unique tunable behavior, emulating real world scenarios where drivers have their own driving styles. Similarly, the maneuver can be performed using different aggressiveness levels, which is particularly useful to manage busy scenarios where conservative rule-based policies would result in undefined waits.