Data-Driven Competitor-Aware Positioning in On-Demand Vehicle Rental Networks

We study a novel operational problem that considers vehicle positioning in on-demand rental networks, such as car sharing in the wider context of a competitive market in which users select vehicles based on access. Existing approaches consider networks in isolation; our competitor-aware model takes supply situations of competing networks into account. We combine online machine learning to predict market-level demand and supply with dynamic mixed integer nonlinear programming. For evaluation, we use discrete event simulation based on real-world data from Car2Go and DriveNow. Our model outperforms conventional models that consider the fleet in isolation by a factor of two in terms of profit improvements. In the case we study, the highest theoretical profit improvements of 7.5% are achieved with a dynamic model. Operators of on-demand rental networks can use our model under existing market conditions to build a profitable competitive advantage by optimizing access for consumers without the need for fleet expansion. Model effectiveness increases further in realistic scenarios of fleet expansion and demand growth. Our model accommodates rising demand, defends against competitors’ fleet expansion, and enhances the profitability of own fleet expansions.

Seamless Vehicle Positioning by Lidar-GNSS Integration: Standalone and Multi-Epoch Scenarios

Accurate and seamless vehicle positioning is fundamental for autonomous driving tasks in urban environments, requiring the provision of high-end measuring devices. Light Detection and Ranging (lidar) sensors, together with Global Navigation Satellite Systems (GNSS) receivers, are therefore commonly found onboard modern vehicles. In this paper, we propose an integration of lidar and GNSS code measurements at the observation level via a mixed measurement model. An Extended Kalman-Filter (EKF) is implemented to capture the dynamic of the vehicle movement, and thus, to incorporate the vehicle velocity parameters into the measurement model. The lidar positioning component is realized using point cloud registration through a deep neural network, which is aided by a high definition (HD) map comprising accurately georeferenced scans of the road environments. Experiments conducted in a densely built-up environment show that, by exploiting the abundant measurements of GNSS and high accuracy of lidar, the proposed vehicle positioning approach can maintain centimeter-to meter-level accuracy for the entirety of the driving duration in urban canyons.

Integrating Vehicle Positioning and Path Tracking Practices for an Autonomous Vehicle Prototype in Campus Environment

This paper presents the implementation of an autonomous electric vehicle (EV) project in the National Taiwan University of Science and Technology (NTUST) campus in Taiwan. The aim of this work was to integrate two important practices of realizing an autonomous vehicle in a campus environment, including vehicle positioning and path tracking. Such a project is helpful to the students to learn and practice key technologies of autonomous vehicles conveniently. Therefore, a laboratory-made EV was equipped with real-time kinematic GPS (RTK-GPS) to provide centimeter position accuracy. Furthermore, the model predictive control (MPC) was proposed to perform the path tracking capability. Nevertheless, the RTK-GPS exhibited some robust positioning concerns in practical application, such as a low update rate, signal obstruction, signal drift, and network instability. To solve this problem, a multisensory fusion approach using an unscented Kalman filter (UKF) was utilized to improve the vehicle positioning performance by further considering an inertial measurement unit (IMU) and wheel odometry. On the other hand, the model predictive control (MPC) is usually used to control autonomous EVs. However, the determination of MPC parameters is a challenging task. Hence, reinforcement learning (RL) was utilized to generalize the pre-trained datum value for the determination of MPC parameters in practice. To evaluate the performance of the RL-based MPC, software simulations using MATLAB and a laboratory-made, full-scale electric vehicle were arranged for experiments and validation. In a 199.27 m campus loop path, the estimated travel distance error was 0.82% in terms of UKF. The MPC parameters generated by RL also achieved a better tracking performance with 0.227 m RMSE in path tracking experiments, and they also achieved a better tracking performance when compared to that of human-tuned MPC parameters.

Vehicle Location Algorithm Based on Federated Learning and Smart Phone in GNSS Low Sampling Rate Scene

In the existing vehicle positioning system based on Global Navigation Satellite System/Inertial Navigation System (GNSS/INS), when the GNSS signal is lost, the error accumulated by using only the INS will damage the positioning accuracy. In order to improve the accuracy, this paper proposed a positioning method based on data-driven and learning models, which utilized distributed data sets to collaboratively construct accurate positioning models through federated fusion algorithms without sacrificing user privacy. In the field scenarios of IID and Non-IID types, this paper compared the performance of INS and the existing two typical methods, DeepSense and PVAUA, it is verified that the two federated learning algorithm models constructed had higher positioning accuracy in different scenarios and different GNSS signal loss durations. The results were analyzed.