This paper studies the problem of indoor zone temperature control in shared work-spaces equipped with heterogeneous heating and cooling sources with the goal of increased energy savings and environment personalization. We consider two scenarios to assess the performance of our control strategies. The first scenario requires time-bound pre-cooling/pre-heating of a shared space in preparation for a scheduled activity (Scenario A). The second scenario considers a cohabited work-space where occupants have different temperature preferences (Scenario B). Utilizing an on-campus smart conference room (SCR) as a test-bed, we use data-driven model learning to establish a relationship between the room’s heating, ventilation and cooling (HVAC) operations and the zone temperatures. Next, we use a model predictive control (MPC)-based approach to achieve a desired average temperature while minimizing power consumption (for Scenario A) and minimize the thermal discomfort experienced by individuals based on their temperature preferences (for Scenario B). The experimental results show that for Scenario A, the proposed control policy can save a significant amount of energy and achieve the desired mean temperature in the space fairly accurately. We further note that for Scenario B, the control scheme can achieve a significant spatial differentiation in temperature towards satisfying the occupants’ thermal preferences.

Speed regulation with simultaneous electric and fuel energy management in electrified aircraft engines is considered. An optimal control approach is formulated and solved that uses a weighted measure of stored energy and fuel consumption as cost function. The semiactive virtual control technique, originally developed for robots with regenerative drives, is suitably extended and used to treat the torque applied by the electromechanical subsystem upon the engine as an independent control input to be optimized. The standard proportional-integral loop used for the primary fuel flow control is maintained for compatibility with industry practice and gain scheduling and limit management logic schemes. A single tunable weight reflects design preferences for fuel or stored electric energy preservation. The magnitude of the weight yields feedback gains which can effectively reverse the direction of power flow between the engine and the electromechanical system, providing a direct way to induce power extraction or injection in transient and steady regimes. In the steady-state, infinite-horizon case, the optimal torque controller is shown to reduce to simple proportional action on the shaft speed plus a bias, paralleling standard linear quadratic tracking with a custom cost function. The paper includes simulation results using the gas turbine model included in NASA’s open-source T-MATS package (Toolbox for the Modeling and Analysis of Thermodynamic Systems) and a representative electromechanical system.

A foot interface may one day control a third arm to assist the hands with a difficult task, but the interface needs to be easy to use. Developing a good foot interface is challenging because of the need to provide support for the leg, allow the user to disengage with the interface without causing unwanted motion, and make it easy for the user to hold a set position. The addition of friction in the interface can enable the device to meet these goals without negatively affecting performance. Although teleoperation is a well explored area of research, relatively little research has been done that examines the effects of friction on the control interface. This paper presents an experiment in which two foot control interfaces are compared. One device uses friction and the other has no added friction, so there is little resistance to motion in any direction. The experiment uses a reaching task and a path-following task to compare the interfaces. The only statistically significant performance differences were that the friction interface reduced the time needed to stop at a target and reduced excess movement when stopping at a target. Also, subjects indicated a preference for the friction interface. The results show that friction can be added to a foot interface to support the device and user and provide some positive gains in performance.

This paper presents a coordinated semi-adaptive approach to closed-loop control of cardiorespiratory state in critically ill patients through the infusion of sedative and opioid drugs. The proposed approach is built upon an upper level multiple drug coordination loop and a lower level semi-adaptive control loop. The coordination loop recursively adjusts the target set points based on the dose-response relationship of a patient estimated by the semi-adaptive control loop, so as to ensure that the target set points become achievable. The semi-adaptive control loop drives the patient state to the target set points while estimating the patient’s dose-response relationship. Hence, the proposed control approach can adjust target set points erroneously specified by caregivers while respond effectively to the need of individual patients. To realize the proposed control approach, we developed (1) a two-input two-output dose-response model of interacting sedative and opioid drugs; (2) a semi-adaptive control algorithm to drive patient state to target set points while selectively estimating high-sensitivity parameters in the dose-response model; and (3) a multiple drug coordination algorithm that reconciles caregiver preference and individual patient’s drug needs. The proposed control approach was evaluated in an example cardiorespiratory control scenario in which cardiac output and respiratory rate are regulated via the infusion of propofol and remifentanil in an in-silico simulation setting. The results show that the coordinated semi-adaptive control could (1) track achievable target set point with robust transient and steady-state error performance and (2) adjust the unachievable target set points to achievable ones.

This paper presents artificial potential field functions to assist adaptive cruise control and cooperative adaptive cruise control vehicles in a multi-lane highway in two dimensions. The artificial field functions consist of three potential fields: road potential, car potential, and velocity potential. The artificial potential field is constructed in a way to assist the vehicles in lane-keeping, speed preference, platoon forming, and obstacle avoidance. The artificial roadway barriers are constructed to keep the car on the highway, and lane boundaries are constructed so that vehicles can overcome the boundary in case lane changing is needed. Also, the shape of the velocity potential is to allow vehicles to either slow down or speed up depending on the surroundings. Static and dynamic obstacles detected by vehicle sensors are superimposed onto these potential fields. Hierarchical control schemes are introduced on top of these inherently reactive potential fields, resulting in stable platoons with short following times. A MATLAB code was generated to show that under our simulated parameters a platoon of vehicles demonstrates good throughput and traffic safety.

For autonomous vehicles to gain widespread customer acceptance, safety and reliability are not nearly enough. Comfort and familiarity of the ride is also of essential importance. Because these are highly subjective factors, autonomous vehicles must be able to adopt personal driving styles to meet individual preference. The adaptive cruise control (ACC) system is a critical function performed by the autonomous vehicle and much research effort has been devoted to the development of a system that acts as a human driver. However, studies which investigate ACC models capable of learning a driving style are limited. In this paper, we propose a method to extract quantifiable parameters which represent a drivers’ driving style and apply these parameters to personalize the longitudinal control of an autonomous vehicle. We then develop a longitudinal driver model that integrates those parameters to enable the ACC system to mimic the driving style of the driver. Finally, the effectiveness of the extraction method and the driver model are obtained through simulation.

This paper presents a novel user interface (UI) for coordinated rate control (CRC) of an excavator end effector using traditional hardware. Coordinated control of an excavator end effector alleviates the cognitive load created by nonlinear arm dynamics on the excavator operator, allowing the operator to perform tasks more quickly and with fewer errors. A human subject experiment demonstrates the feasibility of excavator CRC using the traditional twin joystick setup, and compares operator performance between a CRC UI and traditional excavator UI. Performance of the CRC UI was statically equivalent to the performance of the traditional UI. When asked to self-report UI preference: 26 participants stated they preferred the CRC UI, 6 preferred the traditional UI, and 14 had no preference. Although the current iteration of the CRC UI offered no measurable performance improvements, a remapping of the CRC joystick inputs to the end effector motion could make the CRC UI more intuitive, lead to better performance metrics, and make hydraulic excavators safer, more efficient, and easier to use.

It is well known that obesity, a chronic disease, can lead to increased risk of other serious chronic diseases and even death. Knowledge of daily changes in lean muscle mass and body fat can be helpful in developing personalized diet and exercise routines to correct this problem. In this paper, it is assumed that measurements of individual body composition components are available only periodically although the total body weight is tracked on a daily basis. The control input is physical activity whose profile can be constrained to accommodate individual preferences while the energy intake can be arbitrary. We present switching and time delayed feedback based model-free control methods for the dynamic management of body mass and its major components. Additionally, based on human body weight dynamics, estimation of body composition using soft switching-based observer is proposed. Simulation results validate the performance of the proposed controllers and the observer under disturbances in recording energy intake and energy expenditure figures.

This paper presents systematic analysis and design of power split hybrid configurations using a single planetary gearset and two electric machines for human-powered vehicles. In our design methodology, the cyclist is treated as an integrated part of the powertrain, and the cyclist’s power output is augmented by the battery power, instead of being completely replaced. To obtain the optimal design, all the 12 possible power split hybrid configurations are investigated, and several performance indices, including the cyclist’s oxygen consumption, stamina reduction, and pedaling speed preference, are considered in the optimization problem to evaluate the bicycle design. The dynamic programming technique is used to solve the optimization problem. The optimal design, HyBike-i2, has the pedal connected to the carrier gear and one electric machine to the ring gear on the planetary gearset. The other electric machine and the driven wheel are connected to the sun gear. This design outperforms the normal bicycle (no power assist) and two traditional electric bicycles, and achieves substantial reduction in both the cyclist’s stamina discharge and oxygen consumption when the vehicle operates in the charge-sustaining mode.

When human and robotic agents work together, the challenge in assigning tasks lies in exploiting human strengths, such as expertise and intuition, while still managing the heterogeneous agent team in a near-optimal way. An extension to the Gale-Shapley stable matching algorithm that combines a sequential greedy approach is proposed to apply to task allocation missions. Conventional task features are modeled in the form of task preferences; agent inputs are modeled in the form of agent preferences. The algorithm is applied to a bomb defusal scenario, where bomb location is known but time for each agent to defuse each bomb is supplied through agent preferences. Simulation results are presented, and the sequential greedy Gale-Shapley algorithm is compared to a corresponding sequential single-item auction algorithm under three evaluation criteria — mission completion time, agent-task pair regret, and evenness of task distribution among agents.

This paper studies the response of zebrafish to a bioinspired robotic-fish. The robot’s color pattern and morphophysiology are modulated to emphasize features normally admired by zebrafish in conspecifics. A three-chambered instrumented tank is utilized to conduct a series of preference tests, offering the robotic-fish as a stimulus, juxtaposed with an empty compartment, to live zebrafish. The time spent by fish in the proximity of either stimulus compartment is used to score the preference of an individual to a stimulus. The tail-beating motion of the robotic fish is manipulated based on closed- and open-loop control strategies. The closed-loop controller uses the distance of the live-fish from the robotic-stimulus as its control input, while the open-loop controller provides a tail-beating motion, irrespective of fish behavior. Live-fish locomotion patterns and their preference space are compared and analyzed to ascertain the effects of closed-loop control on zebrafish response. This study’s results suggest that closed-loop control reinforces attraction to the robotic-stimulus, as compared to the open-loop approach, over extended exposure to the robot, therefore aiding against habituation to a stimulus.

Traditional electromyopgrahic (EMG) measurements are based on single sensor information. Due to the arrangement of skeletal muscle fibers for hand motions, cross talk is an inherent problem when inferring motion/force potentials from EMG data. This paper studies means of using sensor arrays to infer better motion/force potential for prosthetic hands. In particular, a surface electromyographic (sEMG) sensor array is used to investigate multiple model fusion techniques. This paper provides a comparison between three statistical model selection criteria. The sEMG signals are pre-processed using four filters, Butterworth, Chebyshev type-II, as well as Bayesian filters such as the Exponential and Half-Gaussian filter. Output Error (OE) models were extracted from sEMG data and hand force data and compared using a Bayesian based fusion model. The four different filters effect were quantified based on the OE models performance in matching the actual measured data. The comparison indicates a preference for using the sensor fusion technique with preprocessed EMG data using the Half-Gaussian Bayesian filter and the Kullback Information Criterion (KIC).

A wind turbine generator offers a green renewable alternative to the traditional fossil and nuclear fuel processes to generate electrical power. Both wind energy technology and wind turbine farm designs remain in demand given the current growth in energy requirements and the public’s preference for clean sources. Simply put, wind energy offers a safe, relatively cost effective solution for global energy production. For example, the energy demand for populated coastal cities encourages offshore farms to fulfill future electrical needs. Similarly, wind turbines may be placed in land-locked regions and power transmitted through electric grids to population centers. Most wind models available to engineers offer superb capabilities for predicting wind velocities on land and far offshore (5 km and greater). However, near shore winds have proven difficult to determine due to surface roughness, thermal stratification, and abrupt displacement height variances. This paper discusses the model comparison of two foremost wind speed prediction tools, AWS Truewinds’ MASS and WindPro’s WAsP. The model comparison is related to measured South Carolina coastal data and suggests AWS Truewind’s MASS wind shear model is the more effective near-shore wind speed prediction tool. In arriving at these results, several areas of future work are discussed.