Search Results

Research interests in autonomous driving have increased significantly in recent years. Several methods are being suggested for performance optimization of autonomous vehicles. However, weather conditions such as rain, snow, and fog may hinder the performance of autonomous algorithms. It is therefore of great importance to study how the performance/efficiency of the underlying scene understanding algorithms vary with such adverse scenarios. Semantic segmentation is one of the most widely used scene-understanding techniques applied to autonomous driving. In this work, we study the performance degradation of several semantic segmentation algorithms caused by rain for off-road driving scenes. Given the limited availability of datasets for real-world off-road driving scenarios that include rain, we utilize two types of synthetic datasets.

There is significant potential for connected and autonomous vehicles to impact vehicle efficiency, fuel economy, and emissions, especially for hybrid-electric vehicles. These improvements could have large-scale impact on oil consumption and air-quality if deployed in large Mobility-as-a-Service or ride-sharing fleets. As part of the US Department of Energy's current Advanced Vehicle Technology Competition (AVCT), EcoCAR: The Mobility Challenge, Mississippi State University’s EcoCAR Team is redesigning and doing the development work necessary to convert a conventional gasoline spark-ignited 2019 Chevy Blazer into a hybrid-electric vehicle with SAE Level 2 autonomy. The target consumer segments for this effort are the Mobility-as-a-Service fleet owners, operators and riders. To accomplish this conversion, the MSU team is implementing a P4 mild hybridization strategy that is expected to result in a 30% increase in fuel economy over the stock Blazer.

The Mississippi State University Formula SAE race car upright was optimized using radial basis function metamodels and an internal state variable (ISV) plasticity damage material model. The weight reduction of the upright was part of a goal to reduce the weight of the vehicle by 25 percent. Using an optimization routine provided an upright design that is lighter that helps to improve vehicle fuel economy, acceleration, and handling. Finite element (FE) models of the upright were produced using quadratic tetrahedral elements. Using tetrahedral elements provided a quick way to produce the multiple FE models of the upright required for the multi-objective optimization. A design of experiments was used to determine how many simulations were required for the optimization. The loads for the simulations included braking, acceleration, and corning loads seen by the car under track conditions.

Although the cost-saving and good environmental impacts are the benefits that make Electric Vehicles (EVs) popular, these advantages are significantly influenced by the cost of battery replacement over the vehicle lifetime. After several charging and discharging cycles, the battery is subjected to energy and power degradation which affects the performance and efficiency of the vehicle. In addition to battery replacement cost, the electricity cost being paid by drivers is another key factor in selecting the EVs. An Energy Management System (EMS) with Model Predictive Control-based (MPC) algorithm is presented for a specific case of heavy-duty EV. Such EV draws its energy from the grid via catenary in addition to the on-board battery. Dynamic model of the vehicle will be defined by State Space Equations (SSE).

A dynamic modeling framework was established to predict status (position, displacement, velocity, acceleration, and shape) of a towed vehicle system with different driver inputs. This framework consists of three components: (1) a state space model to decide position and velocity for the vehicle system based on Newton’s second law; (2) an angular acceleration transferring model, which leads to a hypothesis that the each towed unit follows the same path as the towing vehicle; and (3) a polygon model to draw instantaneous polygons to envelop the entire system at any time point. Input parameters of this model include initial conditions of the system, real-time locations of a reference point (e.g. front center of the towing vehicle) that can be determined from a beacon and radar system, and instantaneous accelerations of this system, which come from driver maneuvers (accelerating, braking, steering, etc.) can be read from a data acquisition system installed on the towing vehicle.

This study examines the performance and subject acceptance level of a hand-operated brake actuator. Using a fixed-base vehicle simulator, data for driver reaction time, stopping time, distance, deceleration, customer acceptance and mental workload were collected. Data for three prototype hand-operated brake actuators and traditional foot-operated brake were compared. An additional test, designed to evaluate anthropometrics, sensitivity, and comfort was performed during training. A user preference survey to determine handbrake acceptance was given to subjects after completing the driving test in the simulator. In certain trials, participants were given the choice of handbrake or footbrake for an unexpected stop condition. When placed into an unexpected braking situation, subjects showed faster brake-application times for operating the hand-operated brake, indicating potential for reduced braking distance.

EcoCAR 2: Plugging In to the Future (EcoCAR) is North America's premier collegiate automotive engineering competition, challenging students with systems-level advanced powertrain design and integration. The three-year Advanced Vehicle Technology Competition (AVTC) series is organized by Argonne National Laboratory, headline sponsored by the U. S. Department of Energy (DOE) and General Motors (GM), and sponsored by more than 28 industry and government leaders. Fifteen university teams from across North America are challenged to reduce the environmental impact of a 2013 Chevrolet Malibu by redesigning the vehicle powertrain without compromising performance, safety, or consumer acceptability. During the three-year program, EcoCAR teams follow a real-world Vehicle Development Process (VDP) modeled after GM's own VDP. The VDP serves as a roadmap for the engineering process of designing, building and refining advanced technology vehicles.

The latest edition of the US Department of Energy's (DOE) Advanced Vehicle Technology Competition (AVTC) series is the EcoCAR Mobility Challenge (EMC). In the third year of the EMC, the Mississippi State University (MSU) team developed and tested a perception system and a longitudinal controller to achieve SAE level 2 autonomy. Our team leveraged the model-based design approach to iterate between developing software components and executing tests in multiple environments in the loop (XIL) to verify that design requirements are met. This workflow allowed us to detect and resolve issues early in the development process. The perception system is composed of a sensor fusion and tracking algorithm. It relies on detections from a front facing camera and radar to generate tracks for a leading vehicle. The tracks from the perception system are used by a model predictive controller (MPC) to maintain a safe distance to the leading vehicle.

The EcoCAR Mobility Challenge (EMC) is the latest edition of the Advanced Vehicle Technology Competition (AVTC) series sponsored by the US Department of Energy. This competition challenges 11 North American universities to redesign a stock 2019 Chevrolet Blazer into an energy-efficient, SAE level 2-autonomous mild hybrid electric vehicle (mHEV) for use in the Mobility as a Service (MaaS) market. The Mississippi State University (MSU) team designed a P4 electric powertrain with an 85kW (113.99 HP) permanent magnet synchronous machine (PMSM) powered by a custom 5.4 kWh lithium-ion energy storage system. To maximize energy efficiency, Model Based Design concepts were leveraged to optimize the overall gear ratio for the P4 system. To accommodate this optimized ratio in the stock vehicle, a custom offset gearbox was designed that links the PMSM to the rear drive module.

The autonomous collision avoidance problem for multi-trailer vehicle maneuvering is investigated in this paper. Different from conventional vehicle systems that contain one single moving part or multi-parts that can be considered as one rigid body, the interconnection between the tractor and each trailer, and interactions between trailers in the multi-trailer system introduce a high dimensional and highly complex dynamic system for the controller design. The external disturbance and parametric uncertainties further increase the difficulty in system identification and state space formulation. To implement a real time control system for various scenarios where the locations and states of the obstacles are not known beforehand, a supervisory algorithm is designed to convert the control problem to a discrete event system. The model predictive control (MPC) using limited lookahead policy is employed in the proposed algorithm.

A robot mining system was developed by the State Space Robotic undergraduate student design team from Mississippi State University (MSU) for the 2016 NASA Robotic Mining Competition. The mining robot was designed to traverse the Martian chaotic terrain, excavate a minimum of 10 kg of Martian regolith and deposit the regolith into a collector bin within 10 minutes as part of the competition. A Systems Engineering approach was followed in proceeding with this design project. The designed mining robot consisted of two major components: (1) mechanical system and (2) control system. This paper mainly focuses on the design and assessment process of the mechanical system but will also briefly mention the control system so as to evaluate the designed robotic system in its entirety. The final designed robot consisted of an aluminum frame driven by four motors and wheels. It utilized a scoop and lifting arm subsystem for collecting and depositing Martian regolith.