Existing robotic arms have limited or no ability to avoid collisions with their environment due mainly to the lack of a suitable sensing system. A collision avoidance capability should be incorporated into every robot so that injuries to people and damage to equipment from collisions are prevented. Important applications that could benefit from robot collision avoidance include: manufacturing, robot-assisted surgery, robotic handling of hazardous waste, and personal robots. Creating a full-coverage, fast, reliable and cost effective sensing system for sensor-based robotic arm collision avoidance is a challenging problem. Capacitive sensors were selected based on their promising potential. Capacitive sensors have the limitations of nonlinearity and being influenced by the environment. In this thesis, their sensing behaviour, and solutions to these limitations, were investigated. A forward model predicts the capacitance for a given electrode geometry. The conventional method, Method of Moments (MoM) and Finite Element Method (FEM) were investigated and compared. The MoM demonstrated that the fringing electric field ignored by the conventional forward model is significant for the robotic arm application due to the relatively large ratio of electrode gap to electrode area. Two forward modeling cases were simulated by writing macro code for a commercial FEM package. The first consisted of two parallel cylindrical robotic arms. The second consisted of two cylindrical shell electrodes wrapped around a pair of robot links that rotated relative to each other. The results for this case were compared with experimental results. The FEM results were a poor predictor of the experimental results. The failure of the FEM model to include the true environmental conditions (e.g. air humidity and surrounding electric fields) is the most likely cause of its inaccuracy. An inverse capacitance model outputs the electrode geometry for a given capacitance. In this research the desired geometric output was the seven robot link pose variables, (x, y, z, q_x, q_y, q_z, q_0), describing the position and the orientation of the link of a robotic arm. A Cerebellar Model Articulation Controller (CMAC) neural network was chosen for the inverse modeling based its ability to model nonlinear behaviour and its efficiency. One CMAC network was trained for each pose variable. The sensor was built using capacitance sensing circuit and a multiplexor board with the potential for 16 by 16 electrode combinations. Note that an n by n combination produces n^2 separate capacitance values. For the inverse modeling experiments, four aluminum foil electrodes were mounted on a CRS-F3 robotic arm and four aluminum foil electrodes were placed on a wooden box used to simulate a second stationary robotic arm. A pair of reference electrodes was mounted on the back of the CRS-F3 arm. This reference measurement was used to normalize the measured capacitances in order to minimize environmental effects. The normalized capacitance data were used to train and test the CMAC neural networks. The CMAC learning factors were dynamically changed to reduce the training errors. A new fuzzy logic approach was developed that allowed the range of the CMAC input data to be increased without significantly increasing the training error. After evaluating eleven combinations of electrodes, it was determined that only the 3 by 3 and 4 by 4 combinations converged with small training errors. Three methods were used to analyze the CMAC testing errors: comparison plots, error plots and error metrics. Over a 15 cm range, pose variable y had maximum absolute errors of 2.1 mm for the 4 by 4 electrode combination and 7.2 mm for the 3 by 3 electrode combination. For the 4 by 4 combination the maximum relative errors were less than 3% for the x, y, and z variables, and less than 15% for the quaternion variables. For the 3 by 3 combination, these values increased to 13% and 20%, respectively. The larger relative errors for the quaternion variables were due to their smaller ranges of variation. Using the same hardware, a simple collision avoidance system was implemented using one pair of electrodes to detect the potential collision between a robotic arm moving in the vertical plane and a second stationary robot. The robot was shown to successfully avoid the potential collision and then continue its motion. / Thesis / Master of Applied Science (MASc)

Collision avoidance is treated as a game of two players with opposing desiderata. In the application to automated car-like vehicles, we will use a differential game in order to model and assess a worst-case analysis. The end result will be an almost analytic representation of a boundary between a "safe" set and a "unsafe" set. We will generalize the research in [27] to non-identical players and begin the setup of the boundary construction. Then we will consider the advantages and disadvantages of manipulation of the cost function through the solution and control techniques. In particular, we introduce a possible way to incorporate a secondary objective such as sticking to a straight path. We also look a hybrid technique to reduce steering when the opposing player is out of the reach of the vehicle; i.e., is out of the "unsafe" set and less extreme maneuvers may be desired. We first look at a terminal cost formulation and through retrograde techniques may shape this boundary between the "safe" and "unsafe" set. We would like this research, or part thereof, to be assessed and simulated on a simulation vehicle such as that used in the Flexible Low-cost Automated Scaled Highway (FLASH) at the Virginia Tech Transportation Institute (VTTI). In preparation, we briefly look at the sensor demands from this game-theoretic approach. / Master of Science

A data structure and update algorithm are presented for a prototype real-time collision avoidance safety system supporting tele-operated robot arms. The data structure is a variant of the octree, which serves as a spatial index. An octree recursively decomposes three dimensional space into eight equal cubic octants (nodes) until each octant meets some decomposition criteria. Our octree stores cylspheres (cylinders with spheres on each end) and rectangular solids as primitives. These primitives make up the two seven-degrees-of-freedom robot arms and environment modeled by the system. Octree nodes containing more than a predetermined number N of primitives are decomposed. This rule keeps the octree small, as the entire environment for our application can be modeled using a few dozen primitives. As robot arms move, the octree is updated to reflect their changed positions. During most update cycles, any given primitive does not change which octree nodes it is in. Thus, modification to the octree is rarely required. Incidents in which one robot arm comes too close to the other arm or· an object in the environment are reported. Cycle time for receiving current joint angles, updating the octree, and detecting/reporting collisions is about 30 milliseconds on an Intel 80386 processor running at 20 MHz. / Master of Science

Many species of bats primarily use echolocation, a type of active sensing wherein bats emit ultrasonic pulses and listen to echoes, for guidance and navigation. Swarms of such bats are a unique type of multi-agent systems that feature bats's echolocation and flight behaviors. In the work of this dissertation, we used bat swarming as an inspiration for multi-agent systems to study various topics which include predation success, active sensing, and collision avoidance. To investigate the predation success, we modeled a group of bats hunting a number of collectively behaving prey. The modeling results demonstrated the benefit of localized grouping of prey in avoiding predation by bats. In the topics regarding active sensing and collision avoidance, we studied individual behavior in swarms as bats could potentially benefit from information sharing while suffering from frequency jamming, i.e., bats having difficulty in distinguishing between self and peers's information. We conducted field experiments in a cave and found that individual bat increased biosonar output as swarm size increased. The experimental finding indicated that individual bat acquired more sensory information in larger swarms even though there could be frequency jamming risk. In a simulation wherein we modeled bats flying through a tunnel, we showed the increasing collision risk in larger swarms for bats either sharing information or flying independently. Thus, we hypothesized that individual bat increased pulse emissions for more sensory information for collision avoidance while possibly taking advantage of information sharing and coping with frequency jamming during swarming. / Ph. D.

The goal of this experiment was to determine what if any effect two different types of warnings have the brake reaction time of distracted commercial motor vehicle operators. The warning conditions were: No Warning, Auditory Tire Skid Warning, and One Second Brake Pulse Warning. Each participant was distracted via a distracter task during the experiment. As the participants were distracted, an obstacle was launched out into their forward path. Each participant received his/her appropriate warning, according to what condition they were placed, when the obstacle entered their headway. It was determined that the Auditory Tire Skid Warning aided in decreasing the movement times, while the One Second Brake Pulse Warning aided in decreasing the number of collisions with the barrels and speed at contact with the barrels. / Master of Science

This thesis presents collision avoidance integrated with lane keeping and adaptive cruise control for a car. Collision avoidance is the ability to avoid obstacles that are in the vehicle's path, without causing damage to the obstacle or car. There are three types of collision avoidance controllers, passive, active, and semi-active. This thesis is designed using active collision avoidance controllers. There are two controllers developed for collision avoidance in this paper. They are an optimal controller and a fuzzy controller. The optimal vehicle trajectory, which maximizes the distance to an obstacle and changes lanes, is derived. The optimal collision avoidance controller is a closed loop controller; with the decisions based on the current state. The fuzzy controller makes decisions based on the system rules. A simulation environment was created to compare these two controllers as viable solutions for collision avoidance. The environment uses MATLAB/Simulink for simulation of the vehicle as well as the optimal and fuzzy controllers. The simulation incorporates system blocks of the kinematics of a car, navigation, states, control law, and velocity controller. Once the controllers are fully developed and tested in the simulation environment, they are implemented and tested on the platform vehicle. This verifies the real world performance of the controllers. The platform vehicle is a modified radio controlled car. This car is completely autonomous. The car has onboard sensors that allow it to follow a white piece of tape as well as detect obstacles. / Master of Science

Presented in this thesis is a collision avoidance algorithm designed around an arc path model. The algorithm was designed for use on Virginia Tech robots entered in the 2003 and 2004 Intelligent Ground Vehicle Competition (IGVC) and on our 2004 entry into the DARPA Grand Challenge. The arc path model was used because of the simplicity of the calculations and because it can accurately represent the base kinematics for Ackerman or differentially steered vehicles. Clothoid curves have been used in the past to create smooth paths with continuously varying curvature, but clothoids are computationally intensive. The circular arc algorithm proposed here is designed with simplicity and versatility in mind. It is readily adaptable to ground vehicles of any size and shape. The algorithm is also designed to run with minimal tuning. The algorithm can be used as a stand alone reactive collision avoidance algorithm in simple scenarios, but it can be better optimized for speed and safety when guided by a global path planner. A complete navigation architecture is presented as an example of how obstacle avoidance can be incorporated in the algorithm. / Master of Science

The autonomous operation of Uncrewed Aircraft Systems (UAS) beyond the pilot in command's visual line of sight is currently restricted due to a lack of cost-effective surveillance sensors robust enough to operate in low-level airspace. The current sensors available either have have high accuracy of locating targets but are too short of a range to be usable or have long ranges but have gaps in coverage due to varying terrain. Sensor fusion is one possible method of combining the strengths of different sensors to increase the overall airspace surveillance quality to allow for robust detect and avoid (DAA) capabilities; enabling beyond visual line of sight operations. This thesis explores some of the current techniques and challenges to use sensor fusion for collision avoidance between crewed aircraft and UAS. It demonstrates an example method of sensor fusion using data from two radars and an ADS-B receiver. In this thesis, a test bed for ground-based airspace monitoring surveillance is proposed for a low cost method of long-term sensor evaluation. Lastly, an potential method of a heterogeneous, score-based, sensor fusion is presented and simulated. / Master of Science / Long range operations of Uncrewed Aircraft Systems (UAS) are currently restricted due to a lack of cost-effective surveillance sensors that work well enough near the ground in the presence changing terrain. The current sensors available either have have high accuracy of locating targets but are too short of a range to be usable or have long ranges but have gaps in coverage due to varying terrain. Sensor fusion is a solution to this problem by combining the strengths of different sensors to allow for better collision avoidance capabilities; enabling these long range operations. This thesis explores some of the current techniques and challenges to use sensor fusion for collision avoidance between crewed aircraft and UAS. It demonstrates an example method of sensor fusion using data from two radars and an ADS-B receiver. In this thesis, a test bed for ground-based airspace monitoring surveillance is proposed for long-term sensor testing. Lastly, an potential method of a sensor fusion using different types of sensors is presented and simulated.

Highly Automated Vehicles (AVs) have the capability to revolutionize the transportation system. These systems have the possibility to make roads safer as AVs do not have limitations that human drivers do, many of which are common causes of vehicle crashes (e.g., distraction or fatigue) often defined generically as human error. The deployment of AVs is likely to be very gradual however, and there will exist situations in which the AV will be driving in close proximity with human drivers across the foreseeable future. Given the persistent crash problem in which the makority of crashes are attributed to driver error, humans will continue to create potential collision scenarios that an AV will be expected to try and avoid or mitigate if developed appropriately. The absence of unreasonable risk in an AVs ability to comprehend and react in these situations is referred to as operational safety. Unlike advanced driver assistance systems (ADAS), highly automated vehicles are required to perform the entirety of the dynamic driving task (DDT) and have a greater responsibility to achieve a high level of operational safety. To address this concern, scenario-based testing has increasingly become a popular option for evaluating AV performance. On a functional level, an AV typically consists of three basic systems: the perception system, the decision and path planning system, and vehicle motion control system. A minimum level of performance is needed in each of these functional blocks to achieve an adequate level of operational safety. The goal of this study was to investigate the effects that perception system performance (i.e., target object state errors) has on vehicle operational safety in collision scenarios similar to that created by human drivers. In the first part of this study, recent annual crash data was used to define a relevant crash population of possible scenarios involving intersections that an AV operating as an urban taxi may encounter. Common crash maneuvers and characteristics were combined to create a set of testing scenarios that represent a high iii percentage of the overall crash population. In the second part of this study, each test scenario was executed using an AV test platform during closed road testing to determine possible real-world perception system performance. This provided a measure of the error in object detection measurements compared to the ideal (i.e., where a vehicle was detected to be compared to where it actually was). In the third part of this study, a set of vehicle simulations were performed to assess the effect of perception system performance on crash outcomes. This analysis simulated hypothetical crashes between an AV and one other collision partner. First an initial worst-case impact configuration was defined and was based on injury outcomes seen in crash data. The AV was then simulated to perform a variety of evasive maneuvers based on an adaptation of a non-impaired driver model. The impact location and orientation of the collision partner was simulated as two states: one based on the object detection of an ideal perception system and the other based on the object detection of the perception system from the AV platform used during the road testing. For simulations in which the two vehicles contacted each other, a planar momentum-impulse model was used for impact modeling and injury outcomes were predicted using an omni-directional injury model taken from recent literature. Results from this study indicate that errors in perception system measurements can change the perceived occupant injury risk within a crash. Sensitivity was found to be dependent on the specific crash type as well as what evasive maneuver is taken. Sensitivities occurred mainly due to changes in the principal direction of force for the crash and the interaction within the injury risk prediction curves. In order to achieve full operational safety, it will likely be important to understand the influence that each functional system (perception, decision, and control) may have on AV performance in these crash scenarios. / Master of Science / Highly Automated Vehicles (AVs) have the capability to revolutionize the transportation system. These systems have the possibility to make roads safer as AVs do not have many of the limitations that human drivers do, many of which are common causes of vehicle crashes (e.g., distraction or fatigue). AVs will be expected to drive alongside human drivers, and so these drivers are likely to continue to be at fault in causing crashes. As part of ensuring safety, AVs will reasonably be expected to try and avoid or help reduce the severity of these crashes. AVs operate using three main systems: the perception system which consists of sensors that see the objects around the AV, the decision and path planning system, which makes decision on what the AV will do, and the vehicle motion control system. Due to the nature of the real-world, these systems may not work exactly as intended which may affect the ability of the AV to react to possible crash scenarios. Because of this, the goal of this study was to investigate the effects that perception system performance (i.e., target object state errors) has on the ability of an AV to react to crash scenarios similar to those created by human drivers. This study first defined crash scenarios using real-world crash data. A real-world perception system was then tested in these scenarios to determine object detection performance. Based on this performance, effects on safety were assessed through vehicle crash simulations. Results from this analysis showed that safety can vary based on both perception system performance and crash scenario. This highlights that it will be important to address system performance in order to achieve high levels of driving safety.

Previous research efforts aimed at understanding the relationship between automation reliability and reliance on the automation have mainly focused on a single dimension of reliability, the automations error rate. Efforts to understand the effects of additional dimensions, such as types of errors, have merely provided suggestions about the effects that automation false alarms and misses can have on human behavior). Furthermore, other dimensions of reliability, such as the distribution of errors in time, have been almost completely ignored. A multi-task simulation of an agricultural vehicle was used in this investigation. The simulator was composed of two main tasks, a collision avoidance task and a tracking task. The collision avoidance task was supported by an imperfect automated collision avoidance system and the tracking task was performed manually. The results of this investigation indicated that there are distinct patterns of reliance that develop as a function of error type, which are dependent on the state of the automation (alarms or non-alarms). The different distributions of errors across time had an effect on the estimates of reliability and subjective trust ratings. The recency of errors was negatively related to perceived reliability and trust. The results of the current investigation also suggest that older adults are able to adjust their behavior according to the characteristics of the automation, although it takes them longer to do so. Furthermore, it appears that older adults are willing to use automated systems, as long as they are reliable enough to reduce workload.