Development and testing of braking and acceleration features for vehicle advanced driver assistance system

<span lang="EN-US">Traffic congestion is a constant problem for cities worldwide. The human driving inefficiency and poor urban planning and development contribute to traffic buildup and travel discomfort. An example of human inefficiency is the phantom traffic jam, which is caused by unnecessary braking, causing traffic to slow down, and eventually coming to a stop. In this study, a brake and acceleration feature (BAF) for the advanced driver assistance system (ADAS) is proposed to mitigate the effects of the phantom traffic phenomenon. In its initial stage, the BAF provides a heads-up display that gives information on how much braking and acceleration input is needed to maintain smooth driving conditions, i.e., without sudden acceleration or deceleration, while observing a safe distance from the vehicle in front. BAF employs a fuzzy logic controller that takes distance information from a light detection and ranging (LIDAR) sensor and the vehicle’s instantaneous speed from the engine control unit (ECU). It then calculates the corresponding percentage value of needed acceleration and braking in order to maintain travel objectives of smooth and safe-distance travel. Empirical results show that the system suggests acceleration and braking values slightly higher than the driver’s actual inputs and can achieve 90% accuracy overall.</span>

Stochastic model predictive control for driver assistance control of intelligent vehicles considering uncertain driving environment

Advanced driver assistance control faces great challenges in cooperating with the nearby vehicles. The assistance controller of an intelligent vehicle has to provide control efforts properly to prevent possible collisions without interfering with the drivers. This paper proposes a novel driver assistance control method for intelligent ground vehicles to cooperate with the nearby vehicles, using the stochastic model predictive control algorithm. The assistance controller is designed to correct the drivers’ steering maneuvers when there is a risk of possible collisions, so that the drivers are not interfered. To enhance the cooperation between the vehicles, the nearby vehicle motion is predicted and included in the assistance controller design. The position uncertainties of the nearby vehicle are considered by the stochastic model predictive control approach via chance constraints. Simulation studies are conducted to validate the proposed control method. The results show that the assistance controller can help the drivers avoid possible collisions with the nearby vehicles and the driving safety can be guaranteed.

iLDM: An Interoperable Graph-Based Local Dynamic Map

Local dynamic map (LDM) is a key component in the future of autonomous and connected vehicles. An LDM serves as a local database with the necessary tools to have a common reference system for both static data (i.e., map information) and dynamic data (vehicles, pedestrians, etc.). The LDM should have a common and well-defined input system in order to be interoperable across multiple data sources such as sensor detections or V2X communications. In this work, we present an interoperable graph-based LDM (iLDM) using Neo4j as our database engine and OpenLABEL as a common data format. An analysis on data insertion and querying time to the iLDM is reported, including a vehicle discovery service function in order to test the capabilities of our work and a comparative analysis with other LDM implementations showing that our proposed iLDM outperformed in several relevant features, furthering its practical utilisation in advanced driver assistance system development.

Approximate reinforcement learning to control beaconing congestion in distributed networks

In vehicular communications, the increase of the channel load caused by excessive periodical messages (beacons) is an important aspect which must be controlled to ensure the appropriate operation of safety applications and driver-assistance systems. To date, the majority of congestion control solutions involve including additional information in the payload of the messages transmitted, which may jeopardize the appropriate operation of these control solutions when channel conditions are unfavorable, provoking packet losses. This study exploits the advantages of non-cooperative, distributed beaconing allocation, in which vehicles operate independently without requiring any costly road infrastructure. In particular, we formulate the beaconing rate control problem as a Markov Decision Process and solve it using approximate reinforcement learning to carry out optimal actions. Results obtained were compared with other traditional solutions, revealing that our approach, called SSFA, is able to keep a certain fraction of the channel capacity available, which guarantees the delivery of emergency-related notifications with faster convergence than other proposals. Moreover, good performance was obtained in terms of packet delivery and collision ratios.

Imaging lidar prototype with homography and deep learning ranging methods

We report on developing a non-scanning laser-based imaging lidar system based on a diffractive optical element with potential applications in advanced driver assistance systems, autonomous vehicles, drone navigation, and mobile devices. Our proposed lidar utilizes image processing, homography, and deep learning. Our emphasis in the design approach is on the compactness and cost of the final system for it to be deployable both as standalone and complementary to existing lidar sensors, enabling fusion sensing in the applications. This work describes the basic elements of the proposed lidar system and presents two potential ranging mechanisms, along with their experimental results demonstrating the real-time performance of our first prototype.