Search Results

Practical applications of recently developed sensor fusion algorithms perform poorly in the real world due to a lack of proper evaluation during development. Existing evaluation metrics do not properly address a wide variety of testing scenarios. This issue can be addressed using proactive performance measurements such as the tools of resilience engineering theory rather than reactive performance measurements such as root mean square error. Resilience engineering is an established discipline for evaluating proactive performance on complex socio-technical systems which has been underutilized for automated vehicle development and evaluation. In this study, we use resilience engineering metrics to assess the performance of a sensor fusion algorithm for vehicle localization. A Kalman Filter is used to fuse GPS, IMU and LiDAR data for vehicle localization in the CARLA simulator.

Standard operation of autonomous vehicles on public roads results in significant exposure to high levels of risk. There is a significant need to develop metrics that evaluate safety of an automated system without reliance on the rate of vehicle accidents and fatalities compared to the number of miles driven; a proactive rather than a reactive metric is needed. Resilience engineering is a new paradigm for safety management that focuses on evaluating complex systems and their interaction with the environment. This paper presents the overall methodology of resilience engineering and the resilience assessment grid (RAG) as an evaluation tool to measure autonomous systems' resilience. This assessment tool was used to evaluate the ability to respond to the system. A Pure Pursuit controller was developed and utilized as the path tracking control algorithm, and the Carla simulator was used to implement the algorithm and develop the testing environment for this methodology.

The transportation sector contributes significantly to emissions and air pollution globally. Emission models of modern vehicles are important tools to estimate the impact of technologies or controls on vehicle emission reductions, but developing a simple and high-fidelity model is challenging due to the variety of vehicle classes, driving conditions, driver behaviors, and other physical and operational constraints. Recent literature indicates that neural network-based models may be able to address these concerns due to their high computation speed and high-accuracy of predicted emissions. In this study, we seek to expand upon this initial research by utilizing several deep neural networks (DNN) architectures such as a recurrent neural network (RNN) and a convolutional neural network (CNN). These DNN algorithms are developed specific to the vehicle-out emissions prediction application, and a comprehensive assessment of their performances is done.

Prediction of vehicle velocity is important since it can realize improvements in the fuel economy/energy efficiency, drivability, and safety. Velocity prediction has been addressed in many publications. Several references considered deterministic and stochastic approaches such as Markov chain, autoregressive models, and artificial neural networks. There are numerous new sensor and signal technologies like vehicle-to-vehicle and vehicle-to-infrastructure communication that can be used to obtain inclusive datasets. Using these inclusive datasets of sensors in deep neural networks, high accuracy velocity predictions can be achieved. This research builds upon previous findings that Long Short-Term Memory (LSTM) deep neural networks provide low error velocity prediction. We developed an LSTM deep neural network that uses different groups of datasets collected in Fort Collins, Colorado.

Government regulations restrict the evaporative emissions during refueling to 0.20 grams per gallon of dispensed fuel. This requires virtually all of the vapors generated and displaced while refueling to be stored onboard the vehicle. The refueling phenomenon of spitback and early-clickoff are also important considerations in designing refueling systems. Spitback is fuel bursting past the nozzle and into the environment and early-clickoff is the pump shutoff mechanism being triggered before the tank is full. Development of a new refueling system design is required for each vehicle as packaging requirements change. Each new design (or redesign) must be prototyped and tested to ensure government regulations and customer satisfaction criteria are satisfied. Often designs need multiple iterations, costing money and time in prototype-based validation procedures. To conserve resources, it is desired to create a Computational Fluid Dynamics (CFD) tool to assist in design validation.

Autonomous vehicle development has benefited from sanctioned competitions dating back to the original 2004 DARPA Grand Challenge. Since these competitions, fully autonomous vehicles have become much closer to significant real-world use with the majority of research focused on reliability, safety and cost reduction. Our research details the recent challenges experienced at the 2017 Self Racing Cars event where a team of international Udacity students worked together over a 6 week period, from team selection to race day. The team’s goal was to provide real-time vehicle control of steering, braking, and throttle through an end-to-end deep neural network. Multiple architectures were tested and used including convolutional neural networks (CNN) and recurrent neural networks (RNN). We began our work by modifying a Udacity driving simulator to collect data and develop training models which we implemented and trained on a laptop GPU.

Vehicle control using prediction based optimal energy management has been demonstrated to achieve better fuel economy resulting in economic, environmental, and societal benefits. However, research focusing on prediction derivation for use in optimal energy management is limited despite the existence of hundreds of optimal energy management research papers published in the last decade. In this work, multiple data sources are used as inputs to derive a prediction for use in optimal energy management. Data sources include previous drive cycle information, current vehicle state, the global positioning system, travel time data, and an advanced driver assistance system (ADAS) that can identify vehicles, signs, and traffic lights. To derive the prediction, the data inputs are used in a nonlinear autoregressive artificial neural network with external inputs (NARX).

High accuracy hybrid vehicle fuel consumption (FC) and emissions models used in practice today are the product of years of research, are physics based, and bear a large computational cost. However, it may be possible to replace these models with a non-physics based, higher accuracy, and computationally efficient versions. In this research, an alternative method is developed by training and testing a time series artificial neural network (ANN) using real world, on-road data for a hydraulic hybrid truck to predict instantaneous FC and emissions. Parameters affecting model fidelity were investigated including the number of neurons in the hidden layer, specific training inputs, dataset length, and hybrid system status. The results show that the ANN model was computationally faster and predicted FC within a mean absolute error of 0-0.1%. For emissions prediction the ANN model had a mean absolute error of 0-3% across CO2, CO, and NOx aggregate predicted concentrations.

The EcoCAR3 competition challenges student teams to redesign a 2016 Chevrolet Camaro to reduce environmental impacts and increase energy efficiency while maintaining performance and safety that consumers expect from a Camaro. Energy management of the new hybrid powertrain is an integral component of the overall efficiency of the car and is a prime focus of Colorado State University’s (CSU) Vehicle Innovation Team. Previous research has shown that error-less predictions about future driving characteristics can be used to more efficiently manage hybrid powertrains. In this study, a novel, real-world implementable energy management strategy is investigated for use in the EcoCAR3 Hybrid Camaro. This strategy uses a Nonlinear Autoregressive Artificial Neural Network with Exogenous inputs (NARX Artificial Neural Network) trained with real-world driving data from a selected drive cycle to predict future vehicle speeds along that drive cycle.

EcoCAR 3 is a university based competition with the goal of hybridizing a 2016 Chevrolet Camaro to increase fuel economy, decrease environmental impact, and maintain user acceptability. To achieve this goal, university teams across North America must design, test, and implement automotive systems. The Colorado State University (CSU) team has designed a parallel pretransmission plug in hybrid electric design. This design will add torque from the engine and motor onto a single shaft to drive the vehicle. Since both the torque generating devices are pre-transmission the torque will be multiplied by both the transmission and final drive. To handle the large amount of torque generated by the entire powertrain system the vehicle's rear half-shafts require a more robust design. Taking advantage of this, the CSU team has decided to pursue the use of composites to increase the shaft's robustness while decreasing component weight.

The design of vehicles, particularly hybrid and other advanced technology vehicles, is typically complex and benefits from systems engineering processes. Vehicle modeling and simulation have become increasingly important system design tools to improve the accuracy, repeatability, and flexibility of the design process. In developing vehicle computational models and simulation, there is an inevitable compromise between the level of detail and the development/computational cost. The tradeoff is specific to the requirements of each vehicle design effort. The assumptions and detail limitations used for vehicle simulations lead to a varying degree of result uncertainty for each design effort. This paper provides a literature review to investigate the state of the art vehicle simulation methods, and quantifies the uncertainty associated with components that are commonly allocated uncertainty.

From November 2010 until May of 2011, NASA's Icing Remote Sensing System was positioned at Platteville, Colorado between the National Science Foundation's S-Pol radar and Colorado State University's CHILL radar (collectively known as FRONT, or ‘Front Range Observational Network Testbed’). This location was also underneath the flight-path of aircraft arriving and departing from Denver's International Airport, which allowed for comparison to pilot reports of in-flight icing. This work outlines how the NASA Icing Remote Sensing System's derived liquid water content and in-flight icing hazard profiles can be used to provide in-flight icing verification and validation during icing and non-icing scenarios with the purpose of comparing these times to profiles of polarized moment data from the two nearby research radars.

There are many design parameters in designing multi-element racecar wings even after the airfoil to be used has been determined. To choose the best parameter values for the wings of a Formula SAE car, Computational Fluid Dynamics combined with highly fractional factorial design of experiments was used. The CFD results were analyzed for the effectiveness of each parameter in increasing the down force, and effective parameters were used for the next CFD analyses with a fractional factorial design for choosing the best parameter values. The designed wings satisfied the target performance criteria.

Motorsport Engineering is developing a foothold, around the World, as a field of academic preparation at the post-graduate level. To gain the appropriate practical skills to augment classroom education, and thus, for the graduates to successfully compete for employment in the Motorsport Industry, it is critical that the degree program has a strong experiential component. This paper describes the need to take an engineering approach to motorsport education by combining a discovery-based education with the traditional lecture format to realize synergistic results. The idea is that to effectively “engineer” the graduate, the student must have a strong skill set or a strong grasp of the fundamentals. The growth of the current educational program at Colorado State University and the effectiveness of merging the “inside-out” process, typical of the research mission, with the instructional practices of the University and with the needs of the Motorsport Industry are discussed.

A study is presented in which light-weight composite materials are used for an engine intake valve. This paper is an interim progress report and documents the successful demonstration of a net-shape, resin transfer molded intake valve in a running engine. A short review of a previous dynamic model is presented showing the advantages in engine performance by using the composite valves. It is shown that the use of reduced mass composite valves allows for increased engine speed and/or more aggressive cam profiles without sacrificing valve strength or stiffness while at the same time maintaining reliable operation. The use of composite materials allows for a significant weight reduction compared to more conventional materials such as steel and titanium. A brief review of the use of composite materials is presented. The development and design process for carbon fiber reinforced valves is discussed.

A student design team at Colorado State University (CSU) has developed an innovative snowmobile to compete in the Clean Snowmobile Challenge 2003 competition. This engine concept was originally developed for the CSC 2002 competition and demonstrated the lowest emissions of any engine that competed that year. The team utilized a 3-cylinder, 594cc, loop-scavenged, two-stroke cycle engine (Arctic Cat ZRT600) and then modified the engine to operate with direct in-cylinder fuel injection using the Orbital OCP air-assisted fuel injection system. This conversion required that the team design and cast new heads for the engine. The direct-injection approach reduced carbon monoxide (CO) emissions by 70% and total hydrocarbon (THC) emissions by 90% from a representative stock snowmobile. An oxidation catalyst was then used to oxidize the remaining CO and THC.