arXiv subjects

Huei Peng

Publications and source records attributed to Huei Peng.

At least 37 records · Page 2Linked to original sources

arXiv2020

Generating Socially Acceptable Perturbations for Efficient Evaluation of Autonomous Vehicles

Deep reinforcement learning methods have been widely used in recent years for autonomous vehicle's decision-making. A key issue is that deep neural networks can be fragile to adversarial attacks or other unseen inputs. In this paper, we address the latter issue: we focus on generating socially acceptable perturbations (SAP), so that the autonomous vehicle (AV agent), instead of the challenging vehicle (attacker), is primarily responsible for the crash. In our process, one attacker is added to the environment and trained by deep reinforcement learning to generate the desired perturbation. The reward is designed so that the attacker aims to fail the AV agent in a socially acceptable way. After training the attacker, the agent policy is evaluated in both the original naturalistic environment and the environment with one attacker. The results show that the agent policy which is safe in the naturalistic environment has many crashes in the perturbed environment.

eess.SY↗

arXiv2020

Safe Reinforcement Learning for Autonomous Vehicles through Parallel Constrained Policy Optimization

Reinforcement learning (RL) is attracting increasing interests in autonomous driving due to its potential to solve complex classification and control problems. However, existing RL algorithms are rarely applied to real vehicles for two predominant problems: behaviours are unexplainable, and they cannot guarantee safety under new scenarios. This paper presents a safe RL algorithm, called Parallel Constrained Policy Optimization (PCPO), for two autonomous driving tasks. PCPO extends today's common actor-critic architecture to a three-component learning framework, in which three neural networks are used to approximate the policy function, value function and a newly added risk function, respectively. Meanwhile, a trust region constraint is added to allow large update steps without breaking the monotonic improvement condition. To ensure the feasibility of safety constrained problems, synchronized parallel learners are employed to explore different state spaces, which accelerates learning and policy-update. The simulations of two scenarios for autonomous vehicles confirm we can ensure safety while achieving fast learning.

cs.LG↗

arXiv2019

Mcity Data Collection for Automated Vehicles Study

The main goal of this paper is to introduce the data collection effort at Mcity targeting automated vehicle development. We captured a comprehensive set of data from a set of perception sensors (Lidars, Radars, Cameras) as well as vehicle steering/brake/throttle inputs and an RTK unit. Two in-cabin cameras record the human driver's behaviors for possible future use. The naturalistic driving on selected open roads is recorded at different time of day and weather conditions. We also perform designed choreography data collection inside the Mcity test facility focusing on vehicle to vehicle, and vehicle to vulnerable road user interactions which is quite unique among existing open-source datasets. The vehicle platform, data content, tags/labels, and selected analysis results are shown in this paper.

cs.CV↗

arXiv2019

Modeling Multi-Vehicle Interaction Scenarios Using Gaussian Random Field

Autonomous vehicles are expected to navigate in complex traffic scenarios with multiple surrounding vehicles. The correlations between road users vary over time, the degree of which, in theory, could be infinitely large, thus posing a great challenge in modeling and predicting the driving environment. In this paper, we propose a method to model multi-vehicle interactions using a stochastic vector field model and apply non-parametric Bayesian learning to extract the underlying motion patterns from a large quantity of naturalistic traffic data. We then use this model to reproduce the high-dimensional driving scenarios in a finitely tractable form. We use a Gaussian process to model multi-vehicle motion, and a Dirichlet process to assign each observation to a specific scenario. We verify the effectiveness of the proposed method on highway and intersection datasets from the NGSIM project, in which complex multi-vehicle interactions are prevalent. The results show that the proposed method can capture motion patterns from both settings, without imposing heroic prior, and hence demonstrate the potential application for a wide array of traffic situations. The proposed modeling method could enable simulation platforms and other testing methods designed for autonomous vehicle evaluation, to easily model and generate traffic scenarios emulating large scale driving data.

cs.RO↗

arXiv2019

Vehicle Energy Dataset (VED), A Large-scale Dataset for Vehicle Energy Consumption Research

We present Vehicle Energy Dataset (VED), a novel large-scale dataset of fuel and energy data collected from 383 personal cars in Ann Arbor, Michigan, USA. This open dataset captures GPS trajectories of vehicles along with their time-series data of fuel, energy, speed, and auxiliary power usage. A diverse fleet consisting of 264 gasoline vehicles, 92 HEVs, and 27 PHEV/EVs drove in real-world from Nov, 2017 to Nov, 2018, where the data were collected through onboard OBD-II loggers. Driving scenarios range from highways to traffic-dense downtown area in various driving conditions and seasons. In total, VED accumulates approximately 374,000 miles. We discuss participant privacy protection and develop a method to de-identify personally identifiable information while preserving the quality of the data. After the de-identification, we conducted case studies on the dataset to investigate the impacts of factors known to affect fuel economy and identify energy-saving opportunities that hybrid-electric vehicles and eco-driving techniques can provide. The case studies are supplemented with a number of examples to demonstrate how VED can be utilized for vehicle energy and behavior studies. Potential research opportunities include data-driven vehicle energy consumption modeling, driver behavior modeling, machine and deep learning, calibration of traffic simulators, optimal route choice modeling, prediction of human driver behaviors, and decision making of self-driving cars. We believe that VED can be an instrumental asset to the development of future automotive technologies. The dataset can be accessed at https://github.com/gsoh/VED.

physics.soc-ph↗

arXiv2018

Developing Robot Driver Etiquette Based on Naturalistic Human Driving Behavior

Automated vehicles can change the society by improved safety, mobility and fuel efficiency. However, due to the higher cost and change in business model, over the coming decades, the highly automated vehicles likely will continue to interact with many human-driven vehicles. In the past, the control/design of the highly automated (robotic) vehicles mainly considers safety and efficiency but failed to address the "driving culture" of surrounding human-driven vehicles. Thus, the robotic vehicles may demonstrate behaviors very different from other vehicles. We study this "driving etiquette" problem in this paper. As the first step, we report the key behavior parameters of human driven vehicles derived from a large naturalistic driving database. The results can be used to guide future algorithm design of highly automated vehicles or to develop realistic human-driven vehicle behavior model in simulations.

cs.RO↗

arXiv2018

Enhancing the performance of a safe controller via supervised learning for truck lateral control

Correct-by-construction techniques, such as control barrier functions (CBFs), can be used to guarantee closed-loop safety by acting as a supervisor of an existing or legacy controller. However, supervisory-control intervention typically compromises the performance of the closed-loop system. On the other hand, machine learning has been used to synthesize controllers that inherit good properties from a training dataset, though safety is typically not guaranteed due to the difficulty of analyzing the associated neural network. In this paper, supervised learning is combined with CBFs to synthesize controllers that enjoy good performance with provable safety. A training set is generated by trajectory optimization that incorporates the CBF constraint for an interesting range of initial conditions of the truck model. A control policy is obtained via supervised learning that maps a feature representing the initial conditions to a parameterized desired trajectory. The learning-based controller is used as the performance controller and a CBF-based supervisory controller guarantees safety. A case study of lane keeping for articulated trucks shows that the controller trained by supervised learning inherits the good performance of the training set and rarely requires intervention by the CBF supervisor

eess.SY↗

arXiv2018

Speed trajectory planning at signalized intersections using sequential convex optimization

An algorithm is developed to optimize vehicle speed trajectory over multiple signalized intersections with known traffic signal information to minimize fuel consumption and travel time, and to meet ride comfort requirements using sequential convex optimization method. A comparison between the proposed method and dynamic programming is carried out to verify its optimality. In addition, vehicle motion during turning is studied because of its significant effect on fuel consumption and travel time.

math.OC↗

arXiv2018

Efficient Mobility-on-Demand System with Ride-Sharing

An algorithm to cluster mobility-on-demand trips considering road network structure is developed in this paper. The benefits of our network partition algorithm are demonstrated in numerical simulations, showing that we can use fewer vehicles and can serve more customers with slightly longer wait time by including predicted future travel demand in trip assignment, compared with the benchmark reactive control policy.

math.OC↗

arXiv2018

Eco-Routing based on a Data Driven Fuel Consumption Model

A nonparametric fuel consumption model is developed and used for eco-routing algorithm development in this paper. Six months of driving information from the city of Ann Arbor is collected from 2,000 vehicles. The road grade information from more than 1,100 km of road network is modeled and the software Autonomie is used to calculate fuel consumption for all trips on the road network. Four different routing strategies including shortest distance, shortest time, eco-routing, and travel-time-constrained eco-routing are compared. The results show that eco-routing can reduce fuel consumption, but may increase travel time. A travel-time-constrained eco-routing algorithm is developed to keep most the fuel saving benefit while incurring very little increase in travel time.

stat.AP↗

arXiv2017

Optimal design of three-planetary-gear power-split hybrid powertrains

Many of today's power-split hybrid electric vehicles (HEVs) utilize planetary gears (PGs) to connect the powertrain elements together. Recent power-split HEVs tend to use two PGs and some of them have multiple modes to achieve better fuel economy and driving performance. Looking to the future, hybrid powertrain technologies must be enhanced to design hybrid light trucks. For light trucks, the need for multi-mode and more PGs is stronger, to achieve the required performance. To systematically explore all the possible designs of multi-mode HEVs with three PGs, an efficient searching and optimization methodology is proposed. All possible clutch topology and modes for one existing configuration that uses three PGs were exhaustively searched. The launching performance is first used to screen out designs that fail to satisfy the required launching performance. A near-optimal and computationally efficient energy management strategy was then employed to identify designs that achieve good fuel economy. The proposed design process successfully identify 8 designs that achieve better launching performance and better fuel economy, while using fewer number of clutches than the benchmark and a patented design.

eess.SY↗

arXiv2017

Analysis of mandatory and discretionary lane change behaviors for heavy trucks

The behaviors of heavy vehicles drivers in mandatory and discretionary lane changes are analyzed in this paper. 640 mandatory and 2,035 discretionary lane change events were extracted from a naturalistic driving database. Variations in gap acceptance and lane change duration were investigated. Statistical analysis showed that mandatory lane changes are more aggressive in gap acceptance and lane change execution than discretionary lane changes. The results can be used for microscopic simulations, and design and evaluation of driver-assistant systems.

eess.SY↗

arXiv2017

Gap Acceptance During Lane Changes by Large-Truck Drivers-An Image-Based Analysis

This paper presents an analysis of rearward gap acceptance characteristics of drivers of large trucks in highway lane change scenarios. The range between the vehicles was inferred from camera images using the estimated lane width obtained from the lane tracking camera as the reference. Six-hundred lane change events were acquired from a large-scale naturalistic driving data set. The kinematic variables from the image-based gap analysis were filtered by the weighted linear least squares in order to extrapolate them at the lane change time. In addition, the time-to-collision and required deceleration were computed, and potential safety threshold values are provided. The resulting range and range rate distributions showed directional discrepancies, i.e., in left lane changes, large trucks are often slower than other vehicles in the target lane, whereas they are usually faster in right lane changes. Video observations have confirmed that major motivations for changing lanes are different depending on the direction of move, i.e., moving to the left (faster) lane occurs due to a slower vehicle ahead or a merging vehicle on the right-hand side, whereas right lane changes are frequently made to return to the original lane after passing.

eess.SY↗

arXiv2017

From the Lab to the Street: Solving the Challenge of Accelerating Automated Vehicle Testing

As automated vehicles and their technology become more advanced and technically sophisticated, evaluation procedures that can measure the safety and reliability of these new driverless cars must develop far beyond existing safety tests. To get an accurate assessment in field tests, such cars would have to be driven millions or even billions of miles to arrive at an acceptable level of certainty - a time-consuming process that would cost tens of millions of dollars. Instead, researchers affiliated with the University of Michigan's Mcity connected and automated vehicle center have developed an accelerated evaluation process that eliminates the many miles of uneventful driving activity to filter out only the potentially dangerous driving situations where an automated vehicle needs to respond, creating a faster, less expensive testing program. This approach can reduce the amount of testing needed by a factor of 300 to 100,000 so that an automated vehicle driven for 1,000 test miles can yield the equivalent of 300,000 to 100 million miles of real-world driving. While more research and development needs to be done to perfect this technique, the accelerated evaluation procedure offers a ground-breaking solution for safe and efficient testing that is crucial to deploying automated vehicles.

cs.HC↗

arXiv2017

Analysis of Unprotected Intersection Left-Turn Conflicts based on Naturalistic Driving Data

Analyzing and reconstructing driving scenarios is crucial for testing and evaluating automated vehicles. This research analyzed left turn / straight-driving conflicts at unprotected intersections by extracting actual vehicle motion data from a naturalistic driving database collected by the University of Michigan. Nearly 7,000 Left turn across path opposite direction (LTAP/OD) events involving heavy trucks and light vehicles were extracted and used to build a stochastic model of such LTAP/OD scenarios. Statistical analysis showed that vehicle type is a significant factor, whereas the change of season seems to have limited influence on the statistical nature of the conflict. The results can be used to build HAV testing environments to simulate the LTAP/OD crash cases in a stochastic manner, which is among the top NHTSA identified priority light-vehicle pre-crash scenarios.

eess.SY↗

arXiv2017

Evaluation of Automated Vehicles Encountering Pedestrians at Unsignalized Crossings

Interactions between vehicles and pedestrians have always been a major problem in traffic safety. Experienced human drivers are able to analyze the environment and choose driving strategies that will help them avoid crashes. What is not yet clear, however, is how automated vehicles will interact with pedestrians. This paper proposes a new method for evaluating the safety and feasibility of the driving strategy of automated vehicles when encountering unsignalized crossings. MobilEye sensors installed on buses in Ann Arbor, Michigan, collected data on 2,973 valid crossing events. A stochastic interaction model was then created using a multivariate Gaussian mixture model. This model allowed us to simulate the movements of pedestrians reacting to an oncoming vehicle when approaching unsignalized crossings, and to evaluate the passing strategies of automated vehicles. A simulation was then conducted to demonstrate the evaluation procedure.

cs.MA↗

arXiv2017

Improving Localization Accuracy in Connected Vehicle Networks Using Rao-Blackwellized Particle Filters: Theory, Simulations, and Experiments

A crucial function for automated vehicle technologies is accurate localization. Lane-level accuracy is not readily available from low-cost Global Navigation Satellite System (GNSS) receivers because of factors such as multipath error and atmospheric bias. Approaches such as Differential GNSS can improve localization accuracy, but usually require investment in expensive base stations. Connected vehicle technologies provide an alternative approach to improving the localization accuracy. It will be shown in this paper that localization accuracy can be enhanced using crude GNSS measurements from a group of connected vehicles, by matching their locations to a digital map. A Rao-Blackwellized particle filter (RBPF) is used to jointly estimate the common biases of the pseudo-ranges and the vehicle positions. Multipath biases, which introduce receiver-specific (non-common) error, are mitigated by a multi-hypothesis detection-rejection approach. The temporal correlation of the estimations is exploited through the prediction-update process. The proposed approach is compared to existing methods using both simulations and experimental results. It was found that the proposed algorithm can eliminate the common biases and reduce the localization error to below 1 meter under open sky conditions.

eess.SY↗

arXiv2017

Accelerated Evaluation of Automated Vehicles in Car-Following Maneuvers

The safety of Automated Vehicles (AVs) must be assured before their release and deployment. The current approach to evaluation relies primarily on (i) testing AVs on public roads or (ii) track testing with scenarios defined in a test matrix. These two methods have completely opposing drawbacks: the former, while offering realistic scenarios, takes too much time to execute; the latter, though it can be completed in a short amount of time, has no clear correlation to safety benefits in the real world. To avoid the aforementioned problems, we propose Accelerated Evaluation, focusing on the car-following scenario. The stochastic human-controlled vehicle (HV) motions are modeled based on 1.3 million miles of naturalistic driving data collected by the University of Michigan Safety Pilot Model Deployment Program. The statistics of the HV behaviors are then modified to generate more intense interactions between HVs and AVs to accelerate the evaluation procedure. The Importance Sampling theory was used to ensure that the safety benefits of AVs are accurately assessed under accelerated tests. Crash, injury and conflict rates for a simulated AV are simulated to demonstrate the proposed approach. Results show that test duration is reduced by a factor of 300 to 100,000 compared with the non-accelerated (naturalistic) evaluation. In other words, the proposed techniques have great potential for accelerating the AV evaluation process.

cs.OH↗