arXiv subjects

Xinzhou Wu

Publications and source records attributed to Xinzhou Wu.

6 recordsLinked to original sources

arXiv2021

Combining Reinforcement Learning with Model Predictive Control for On-Ramp Merging

We consider the problem of designing an algorithm to allow a car to autonomously merge on to a highway from an on-ramp. Two broad classes of techniques have been proposed to solve motion planning problems in autonomous driving: Model Predictive Control (MPC) and Reinforcement Learning (RL). In this paper, we first establish the strengths and weaknesses of state-of-the-art MPC and RL-based techniques through simulations. We show that the performance of the RL agent is worse than that of the MPC solution from the perspective of safety and robustness to out-of-distribution traffic patterns, i.e., traffic patterns which were not seen by the RL agent during training. On the other hand, the performance of the RL agent is better than that of the MPC solution when it comes to efficiency and passenger comfort. We subsequently present an algorithm which blends the model-free RL agent with the MPC solution and show that it provides better trade-offs between all metrics -- passenger comfort, efficiency, crash rate and robustness.

cs.RO↗

arXiv2017

An End-to-End System for Crowdsourced 3d Maps for Autonomous Vehicles: The Mapping Component

Autonomous vehicles rely on precise high definition (HD) 3d maps for navigation. This paper presents the mapping component of an end-to-end system for crowdsourcing precise 3d maps with semantically meaningful landmarks such as traffic signs (6 dof pose, shape and size) and traffic lanes (3d splines). The system uses consumer grade parts, and in particular, relies on a single front facing camera and a consumer grade GPS. Using real-time sign and lane triangulation on-device in the vehicle, with offline sign/lane clustering across multiple journeys and offline Bundle Adjustment across multiple journeys in the backend, we construct maps with mean absolute accuracy at sign corners of less than 20 cm from 25 journeys. To the best of our knowledge, this is the first end-to-end HD mapping pipeline in global coordinates in the automotive context using cost effective sensors.

cs.RO↗

arXiv2016

Inter-Vehicle Range Estimation from Periodic Broadcasts

Dedicated short-range communication (DSRC) enables vehicular communication using periodic broadcast messages. We propose to use these periodic broadcasts to perform inter-vehicle ranging. Motivated by this scenario, we study the general problem of precise range estimation between pairs of moving vehicles using periodic broadcasts. Each vehicle has its own independent and unsynchronized clock, which can exhibit significant drift between consecutive periodic broadcast transmissions. As a consequence, both the clock offsets and drifts need to be taken into account in addition to the vehicle motion to accurately estimate the vehicle ranges. We develop a range estimation algorithm using local polynomial smoothing of the vehicle motion. The proposed algorithm can be applied to networks with arbitrary number of vehicles and requires no additional message exchanges apart from the periodic broadcasts. We validate our algorithm on experimental data and show that the performance of the proposed approach is close to that obtained using unicast round-trip time ranging. In particular, we are able to achieve sub-meter ranging accuracies in vehicular scenarios. Our scheme requires additional timestamp information to be transmitted as part of the broadcast messages, and we develop a novel timestamp compression algorithm to minimize the resulting overhead.

cs.NI↗

arXiv2016

Secure Content Distribution in Vehicular Networks

Dedicated short range communication (DSRC) relies on secure distribution to vehicles of a certificate revocation list (CRL) for enabling security protocols. CRL distribution utilizing vehicle-to-vehicle (V2V) communications is preferred to an infrastructure-only approach. One approach to V2V CRL distribution, using rateless coding at the source and forwarding at vehicle relays is vulnerable to a pollution attack in which a few malicious vehicles forward incorrect packets which then spread through the network leading to denial-of-service. This paper develops a new scheme called Precode-and-Hash that enables efficient packet verification before forwarding thereby preventing the pollution attack. In contrast to rateless codes, it utilizes a fixed low-rate precode and random selection of packets from the set of precoded packets. The fixed precode admits efficient hash verification of all encoded packets. Specifically, hashes are computed for all precoded packets and sent securely using signatures. We analyze the performance of the Precode-and-Hash scheme for a multi-hop line network and provide simulation results for several schemes in a more realistic vehicular model.

cs.IT↗

arXiv2015

Distributed Rate and Power Control in Vehicular Networks

The focus of this paper is on the rate and power control algorithms in Dedicated Short Range Communication (DSRC) for vehicular networks. We first propose a utility maximization framework by leveraging the well-developed network congestion control, and formulate two subproblems, one on rate control with fixed transmit powers and the other on power control with fixed rates. Distributed rate control and power control algorithms are developed to solve these two subproblems, respectively, and are proved to be asymptotically optimal. Joint rate and power control can be done by using the two algorithms in an alternating fashion. The performance enhancement of our algorithms compared with a recent rate control algorithm, called EMBARC, is evaluated by using the network simulator ns2.

eess.SY↗

arXiv2012

Distributed Power Control and Coding-Modulation Adaptation in Wireless Networks using Annealed Gibbs Sampling

In wireless networks, the transmission rate of a link is determined by received signal strength, interference from simultaneous transmissions, and available coding-modulation schemes. Rate allocation is a key problem in wireless network design, but a very challenging problem because: (i) wireless interference is global, i.e., a transmission interferes all other simultaneous transmissions, and (ii) the rate-power relation is non-convex and non-continuous, where the discontinuity is due to limited number of coding-modulation choices in practical systems. In this paper, we propose a distributed power control and coding-modulation adaptation algorithm using annealed Gibbs sampling, which achieves throughput optimality in an arbitrary network topology. We consider a realistic Signal-to-Interference-and-Noise-Ratio (SINR) based interference model, and assume continuous power space and finite rate options (coding-modulation choices). Our algorithm first decomposes network-wide interference to local interference by properly choosing a "neighborhood" for each transmitter and bounding the interference from non-neighbor nodes. The power update policy is then carefully designed to emulate a Gibbs sampler over a Markov chain with a continuous state space. We further exploit the technique of simulated annealing to speed up the convergence of the algorithm to the optimal power and coding-modulation configuration. Finally, simulation results demonstrate the superior performance of the proposed algorithm.

cs.NI↗