book/book chapter
published journal article
There is considerable evidence that evaluating the subjective risk level of driving decisions can improve the safety of Autonomous Driving Systems (ADS) in both typical and complex driving scenarios. In this paper, we propose a novel data-driven approach that uses scene-graphs as intermediate representations for modeling the subjective risk of driving maneuvers. Our approach includes a Multi-Relation Graph Convolution Network, a Long-Short Term Memory Network, and attention layers. To train our model, we formulate subjective risk assessment as a supervised scene classification problem. We evaluate our model on both synthetic lane-changing datasets and real-driving datasets with various driving maneuvers. We show that our approach achieves a higher classification accuracy than the state-of-the-art approach on both large (96.4% vs. 91.2%) and small (91.8% vs. 71.2%) lane-changing synthesized datasets, illustrating that our approach can learn effectively even from small datasets. We also show that our model trained on a lane-changing synthesized dataset achieves an average accuracy of 87.8% when tested on a real-driving lane-changing dataset. In comparison, the state-of-the-art model trained on the same synthesized dataset only achieved 70.3% accuracy when tested on the real-driving dataset, showing that our approach can transfer knowledge more effectively. Moreover, we demonstrate that the addition of spatial and temporal attention layers improves our model’s performance and explainability. Finally, our results illustrate that our model can assess the risk of various driving maneuvers more accurately than the state-of-the-art model (86.5% vs. 58.4%, respectively).
published journal article
We consider the problem of collectively transporting multiple objects using air–ground multirobot teams. The objective is to find the optimal matching between the objects and aerial/ground robots that minimizes the energy of the overall system. We reveal the local optimality criteria for this combinatorial problem and prove that combining a branch and bound algorithm with a negative-cycle canceling algorithm (NCCA) yields an efficient algorithm that provides the globally optimal solution of the problem. Numerical experiments demonstrate the performance on practical problems.
policy brief
Workers in Southern California currently face transportationrelated challenges accessing employment opportunities, including but not limited to high parking costs and/or limited parking availability in dense employment and residential areas; long commute distances between residential areas and employment opportunities; and poor transit service quality in many areas. These challenges are particularly burdensome for low-income households that may not have access to a personal vehicle and/or live in jobpoor neighborhoods, as having a personal vehicle may be the only viable way to get to work.
MS Thesis
published journal article
In this article, we implement an estimation procedure for a particular mathematical programming activity-based model to estimate the relative importance of factors associated with spatial and temporal interrelationships among the out-of-home activities that motivate a household’s need or desire to travel. The method uses a genetic algorithm to estimate coefficient values of the utility function, based on a particular multidimensional sequence alignment method to deal with the nominal, discrete attributes of the activity/travel pattern (e.g., which household member performs which activity, which vehicle is used, sequencing of activities), and a time sequence alignment method to handle temporal attributes of the activity pattern (e.g., starting and ending time of each activity and/or travel). The estimation procedure is tested on data drawn from a well-known activity/travel survey.
published journal article |
Jul 2008
published journal article |
Dec 2009
working paper |
Jan 2008
published journal article
research report
This research develops and tests, via microscopic simulation, a real-time adaptive control system for corridor management in the form of three real-time adaptive control strategies: intersection control, ramp control and an integrated control that combines both intersection and ramp control. The development of these strategies is based on a mathematical representation that describes the behavior of traffic flow in corridor networks and actuated controller operation. Only those parameters commonly found in modern actuated controllers (e.g., Type 170 and 2070 controllers) are considered in the formulation of the optimal control problem. As a result, the proposed strategies easily could be implemented with minimal adaptation of existing field devices and the software that controls their operation. Microscopic simulation was employed to test and evaluate the performance of the proposed strategies in a calibrated network. Simulation results indicate that the proposed strategies are able to increase overall system performance and also the local performance on ramps and intersections. Prior to testing the complete model, separate tests were conducted to evaluate the intersection control model on: 1) an isolated intersection, and 2) a network of intersections along an arterial. The complete model was then tested and evaluated on the Alton Parkway/I-405 corridor network in Irvine, California. In testing the optimal control model, we simulated a variety of conditions on the freeway and arterial subsystems that cover the range of demand from peak to non-peak, incident to non-incident, conditions. The results of these experiments were evaluated against full-actuated operation and found to offer improved performance.
research report |
Oct 2008
conference paper
Cyber-physical systems (CPS) play a pivotal role in modern critical infrastructure, spanning sectors such as energy, transportation, healthcare, and manufacturing. These systems combine digital and physical elements, making them susceptible to a new class of threats known as cyber kinetic vulnerabilities. Such vulnerabilities can exploit weaknesses in the cyber world to force physical consequences and pose significant risks to both human safety and infrastructure integrity. This paper presents a novel tool, named Rampo, that can perform binary code analysis to identify cyber kinetic vulnerabilities in CPS. The proposed tool takes as input a Signal Temporal Logic (STL) formula that describes the kinetic effect—i.e., the behavior of the “physical” system—that one wants to avoid. The tool then searches the possible “cyber” trajectories in the binary code that may lead to such “physical” behavior. This search integrates binary code analysis tools and hybrid systems falsification tools using a Counter-Example Guided Abstraction Refinement (CEGAR) approach. In particular, Rampo starts by analyzing the binary code to extract symbolic constraints that represent the different paths in the code. These symbolic constraints are then passed to a Satisfiability Modulo Theories (SMT) solver to extract the range of control signals that can be produced by each of the paths in the code. The next step is to search over possible “physical” trajectories using a hybrid systems falsification tool that adheres to the behavior of the “cyber” paths and yet leads to violations of the STL formula. Since the number of “cyber” paths that need to be explored increases exponentially with the length of “physical” trajectories, we iteratively perform refinement of the “cyber” path constraints based on the previous falsification result and traverse the abstract path tree obtained from the control program to explore the search space of the system. To illustrate the practical utility of binary code analysis in identifying cyber kinetic vulnerabilities, we present case studies from diverse CPS domains, showcasing how they can be discovered in their control programs. In particular, compared to off-the-shelf tools, our tool could compute the same number of vulnerabilities while leading to a speedup that ranges from 3× to 98×.
conference paper
Fast incident response and management are main tasks of Traffic Management Center (TMC) operators. Using the latest microscopic simulation modeling techniques, a TMC operator training system was developed and set up at California Advanced Transportation Management Systems (ATMS) Testbed of University of California, Irvine. The system is designed to duplicate the standardized TMC software systems and data feeds in a secure environment, where TMC operators can be trained to enrich their experiences and enhance their skills based on various incident scenarios. Since the completion of the prototype system, three training classes have been performed and satisfactory results have been obtained. For the covering abstract see ITRD E140665.