Search for: All records

Modern cities have hundreds to thousands of traffic cameras distributed across them, many of them with the capa- bility to pan and tilt, but very often these pan and tilt cameras either do not have angle sensors or do not provide camera orientation feedback. This makes it difficult to robustly track traffic using these cameras. Several methods to automatically detect the camera pose have been proposed in literature, with the most popular and robust being deep learning-based approaches. However, they are compute intensive, require large amounts of training data, and generally cannot be run on embedded devices. In this paper, we propose TIPAngle – a Siamese neural network, lightweight training, and a highly optimized inference mechanism and toolset to estimate camera pose and thereby improve traffic tracking even when operators change the pose of the traffic cameras. TIPAngle is 18.45x times faster and 3x more accurate in determining the angle of a camera frame than a ResNet-18 based approach. We deploy TIPAngle to a Raspberry Pi CPU and show that processing an image takes an average of .057s, equating to a frequency of about 17Hz on an embedded device. 

Prior art in traffic incident detection relies on high sensor coverage and is primarily based on decision-tree and random forest models that have limited representation capacity and, as a result, cannot detect incidents with high accuracy. This paper presents IncidentNet - a novel approach for classifying, localizing, and estimating the severity of traffic incidents using deep learning models trained on data captured from sparsely placed sensors in urban environments. Our model works on microscopic traffic data that can be collected using cameras installed at traffic intersections. Due to the unavailability of datasets that provide microscopic traffic details and traffic incident details simultaneously, we also present a methodology to generate a synthetic microscopic traffic dataset that matches given macroscopic traffic data. IncidentNet1 achieves a traffic incident detection rate of 98%, with false alarm rates of less than 7% in 197 seconds on average in urban environments with cameras on less than 20% of the traffic intersections. 

5G Millimeter Wave (mmWave) technology holds great promise for Connected Autonomous Vehicles (CAVs) due to its ability to achieve data rates in the Gbps range. However, mmWave suffers from a high beamforming overhead and requirement of line of sight (LOS) to maintain a strong connection. For Vehicle-to-Infrastructure (V2I) scenarios, where CAVs connect to roadside units (RSUs), these drawbacks become apparent. Because vehicles are dynamic, there is a large potential for link blockages. These blockages are detrimental to the connected applications running on the vehicle, such as cooperative perception and remote driver takeover. Existing RSU selection schemes base their decisions on signal strength and vehicle trajectory alone, which is not enough to prevent the blockage of links. Many modern CAVs motion planning algorithms routinely use other vehicle’s near-future path plans, either by explicit communication among vehicles, or by prediction. In this paper, we make use of the knowledge of other vehicle’s near future path plans to further improve the RSU association mechanism for CAVs. We solve the RSU association algorithm by converting it to a shortest path problem with the objective to maximize the total communication bandwidth. We evaluate our approach, titled B-AWARE, in simulation using Simulation of Urban Mobility (SUMO) and Digital twin for self-dRiving Intelligent VEhicles (DRIVE) on 12 highway and city street scenarios with varying traffic density and RSU placements. Simulations show B-AWARE results in a 1.05× improvement of the potential datarate in the average case and 1.28× in the best case vs. the state-of-the-art. But more impressively, B-AWARE reduces the time spent with no connection by 42% in the average case and 60% in the best case as compared to the state-of-the-art methods. This is a result of B-AWARE reducing nearly 100% of blockage occurrences. 

A major challenge in cooperative sensing is to weight the measurements taken from the various sources to get an accurate result. Ideally, the weights should be inversely proportional to the error in the sensing information. However, previous cooperative sensor fusion approaches for autonomous vehicles use a fixed error model, in which the covariance of a sensor and its recognizer pipeline is just the mean of the measured covariance for all sensing scenarios. The approach proposed in this paper estimates error using key predictor terms that have high correlation with sensing and localization accuracy for accurate covariance estimation of each sensor observation. We adopt a tiered fusion model consisting of local and global sensor fusion steps. At the local fusion level, we add in a covariance generation stage using the error model for each sensor and the measured distance to generate the expected covariance matrix for each observation. At the global sensor fusion stage we add an additional stage to generate the localization covariance matrix from the key predictor term velocity and combines that with the covariance generated from the local fusion for accurate cooperative sensing. To showcase our method, we built a set of 1/10 scale model autonomous vehicles with scale accurate sensing capabilities and classified the error characteristics against a motion capture system. Results show an average and max improvement in RMSE when detecting vehicle positions of 1.42x and 1.78x respectively in a four-vehicle cooperative fusion scenario when using our error model versus a typical fixed error model. 

Careful placement of a distributed computational application within a target device cluster is critical for achieving low application completion time. The problem is challenging due to its NP-hardness and combinatorial nature. In recent years, learning-based approaches have been proposed to learn a placement policy that can be applied to unseen applications, motivated by the problem of placing a neural network across cloud servers. These approaches, however, generally assume the device cluster is fixed, which is not the case in mobile or edge computing settings, where heterogeneous devices move in and out of range for a particular application. To address the challenge of scaling to different-sized device clusters and adapting to the addition of new devices, we propose a new learning approach called GiPH, which learns policies that generalize to dynamic device clusters via 1) a novel graph representation gpNet that efficiently encodes the information needed for choosing a good placement, and 2) a scalable graph neural network (GNN) that learns a summary of the gpNet information. GiPH turns the placement problem into that of finding a sequence of placement improvements, learning a policy for selecting this sequence that scales to problems of arbitrary size. We evaluate GiPH with a wide range of task graphs and device clusters and show that our learned policy rapidly finds good placements for new problem instances. GiPH finds placements that achieve up to 30.5% better makespan, searching up to 3× faster than other search-based placement policies. 

Different types of lane-changing assistance systems are usually developed separately by different automotive makers or suppliers. A lane-changing model can meet its own requirements, but it may be incompatible with another lane-changing model. In this paper, we verify if two lane-changing models are compatible so that the two corresponding vehicles on different lanes can exchange their lanes successfully. We propose a methodology and an algorithm to perform the verification on the combinations of four lane-changing models. Experimental results demonstrate the compatibility (or incompatibility) between the models. The verification results can be utilized during runtime to prevent incompatible vehicles from entering a lane-changing road segment. To the best of our knowledge, this is the first work considering the compatibility issue for lane-changing models. 

Many Cyber-Physical Systems (CPS) have timing constraints that must be met by the cyber components (software and the network) to ensure safety. It is a tedious job to check if a CPS meets its timing requirement especially when they are distributed and the software and/or the underlying computing platforms are complex. Furthermore, the system design is brittle since a timing failure can still happen e.g., network failure, soft error bit flip, etc. In this paper, we propose a new design methodology called Plan B where timing constraints of the CPS are monitored at the runtime, and a proper backup routine is executed when a timing failure happens to ensure safety. We provide a model on how to express the desired timing behavior using a set of timing constructs in a C/C++ code and how to efficiently monitor them at the runtime. We showcase the effectiveness of our approach by conducting experiments on three case studies: 1) the full software stack for autonomous driving (Apollo), 2) a multi-agent system with 1/10th scale model robots, and 3) a quadrotor for search and rescue application. We show that the system remains safe and stable even when intentional faults are injected to cause a timing failure. We also demonstrate that the system can achieve graceful degradation when a less extreme timing failure happens.