Angel Flores-Abad

Control of a space robot for minimal attitude disturbance to the base satellite for capturing a tumbling satellite

The use of a space manipulator (robot) for capturing a tumbling object is a risky and challenging task, mainly because when the manipulator onboard a servicing satellite (base satellite) intercepts with an external object for capture, the resulting impulse will be transferred along the mechanical arm down to the servicing satellite causing disturbance to the attitude of the satellite. Such disturbance may destabilize the servicing satellite if the captured object is tumbling and the physical contact between the robot end-effector and the object is not controlled properly. Certainly, the risk may be mitigated with a force or impedance control capability of the manipulator. However, the implementation of force or impedance control usually requires the robot to have a joint torque sensing and control capability which is a very expensive requirement for a space manipulator. To date, there has never been a really flown space manipulator having a joint torque control capability. Further, even a force or impedance control capability becomes available, much development is still needed before safe capture of a tumbling object can be confidently tried in a real mission. This paper presents an optimal control strategy for a space manipulator to have minimal impact to the base satellite during a capturing operation. The idea is to first predict an optimal future time and motion state for capturing and then control the manipulator to reach the determined motion state such that, when the tip of the robot maneuvers to and intercepts with the tumbling object, a minimal attitude disturbance to the servicing satellite will occur. The proposed control strategy can be implemented regardless whether the manipulator has a joint torque control capability or not. Since the control acts before a physical contact happens, it will not affect but actually augment any existing force or impedance control capability of the manipulator. The proposed method is demonstrated using a simulation example.

Optimal Capture of a Tumbling Object in Orbit Using a Space Manipulator

This paper introduces an optimal capture strategy for a manipulator based on a servicing spacecraft to approach an arbitrarily rotating object, such as a malfunctioning satellite or a piece of orbital debris, for capturing with minimal impact to the robot’s base spacecraft. The method consists of two steps. The first step is to determine an optimal future time and the target object’s corresponding motion state for the robot to capture the tumbling object, so that, at the time when the gripper of the robot intercepts the target the very first instant, the resulting impact or disturbance to the attitude of the base spacecraft will be minimal. The second step is to control the robot to reach the tumbling object at the predicted optimal time along an optimal trajectory. The optimal control problem is solved with random uncertainties in the initial and final boundary conditions. Uncertainties are introduced because sensor and estimation errors inevitably exist in the first step, i.e., determination process of the initial and final boundary conditions. The application of the method is demonstrated using a dynamics simulation example.

Development of a Small UAV with Autopilot Capability

It is challenging to develop and test autonomous unmanned aerial vehicles (UAVs) because of the multidisciplinary nature of the work and the paramount safety requirement. A UAV system with autopilot capability was developed by a group of students at New Mexico State University. The UAV system was built based on an originally radio controlled (RC) Raptor 90 helicopter and a commercial-off-the-shelf autopilot package MP2128 HELI . The development work was focused on the components selection, interfaces design, system integration and testing. Several problems regarding vibration isolation, electromagnetic interference shielding and safe testing of the system have been solved. This paper describes the development work including the system architecture, mechanical and electrical components design, system integration, and some preliminary results of the system test.

Optimal Control of a Space Robot to Approach a Tumbling Object for Capture with Uncertainties in the Boundary Conditions

This paper presents an optimal control strategy for a space robot to approach a tumbling object, such as an out-of-control satellite or a piece of space debris, for capture with minimal impact to the base satellite (also called servicing satellite or chaser satellite) with consideration of random uncertainties in the initial and final boundary conditions. The method consists of two steps. The first step is to determine an optimal future time and the target object’s corresponding motion state for the robot to capture the tumbling object, such that, at the time when the tip of the robot intercepts it, the resulting impact or disturbance on the attitude of the base satellite will be minimal. In the second step, the space robot will be controlled to reach the tumbling object at the predicted optimal time along an optimal trajectory. Uncertainties in the initial and final boundary conditions are introduced as errors inevitably exist in the tracking sensing data. Markov Chain Monte Carlo (MCMC) method is employed to solve the optimal control problem with boundary uncertainties. The performance of the method is demonstrated using a dynamics simulation example.

Bio-inspired approach for a space manipulator to capture a tumbling object with minimal impact force

This paper presents a bio-inspired methodology aimed at achieving minimal impact force to the robot and the base satellite during a robotic capturing task for a satellite on-orbit servicing mission. First, from the assumed visual observation of the target satellite’s motion, an optimal time and target configuration for the capturing operation are obtained. The optimal time is a future time such that, when the robot touches the tumbling object, there will be zero or minimal attitude disturbance in the base satellite. Second, tau theory is employed to plan an end-effector motion trajectory which will be used to guide the robot to reach the target at the optimal time. Tau theory is utilised because it has been found to be the natural way that a human arm and some animals use to reach an object for capture, thus it has been optimised through the natural evolution and is worthy of study. Two simulation examples are given to show the performance of the presented methodology and comparisons with a traditional path planning method are discussed. A comparison of the method with a traditional path planning method is also discussed.

Development of a Special Inertial Measurement Unit for UAV Applications

Dynamics modeling is becoming more and more important in the development and control of unmanned aerial vehicles (UAV). An accurate model of a vehicle requires good knowledge of the dynamics properties and motion states, which are usually estimated with the help of integrated inertial measurement units (IMUs). This work develops a special six degrees of freedom IMU, which has the capability of measuring the angular accelerations. This paper introduces the design of the new IMU along with its sensor models and calibration procedures. The work introduces two experimental methods to verify the calibrated IMU readings. The IMU was designed to support an on-line methodology to estimate the parameters of UAV’s dynamics model that is currently being developed by the authors. [DOI: 10.1115/1.4007122]

Test of a Special Inertial Measurement Unit using a Quadrotor Aircraft

This paper discusses the calibration and test of a special Inertial Measurement Unit (IMU) using a Quadrotor aircraft. The IMU was designed and built in house, which has the capability of estimating the angular acceleration in addition to the measurement of angular velocity and linear accelerations. Among the different types of existing UAVs, a Quadrotor was chosen as a platform to test the IMU because its simplicity in geometry structure and control. Before the IMU can be practically used for real UAV flight control, a full 6-DOF flight test must be successfully done. This work reported in this paper addresses this need. For safety, the flight tests are performed in a gravity-balanced test-stand which was also designed in house for facilitating safe test of micro UAVs.

Development of an autonomous unmanned aerial vehicle using gas-powered RC helicopter

This paper describes the development of a small unmanned aerial vehicle (UAV) with autonomous flight capability. It is the result of integrating a commercial off-the-shelf autopilot system with a low-cost gas-powered RC helicopter. Researchers face several challenges when developing UAVs using gas-powered RC helicopters. To avoid the corruption of the structural vibration to the avionics hardware system, an innovative vibration isolation technique is developed to mechanically isolate the vibrations from the gas engine and rotors. Furthermore, a dynamics model for the vehicle is created to support the autonomous flight control development work. Further, a loss-recover method using Kalman filter is employed for estimating the attitude and position statuses when GPS signal is lost. Several other key challenges related to electromagnetic interference shielding and safe flight testing are also effectively solved in the project. The integration work has been completed and the test flights done so far show that the developed autonomous UAV works well under the integration of the mechanical system, electronic system, and controller software.

A Robotic Concept for the NASA Asteroid-capture Mission

This paper proposes an approach for capturing and despinning a tumbling asteroid using a space robot. The capturing and despining approach consists of four phases. In the first phase, the capturing system is controlled to match the linear and angular velocities of the asteroid while the path of its center of mass (CM) is made to follow the path of the asteroid’s CM. The second phase is divided into two subphases. In the first subphase, a robotic manipulator is deployed until a camera mounted at the end effector (EE) faces the target landing location. During the second phase, the EE of manipulator approaches a chosen point on the surface of the asteroid. Attitude controllers driving the thrusters at the base of the spacecraft ensure that the matching velocity errors remain bounded while the robotic arm is deployed. Cameras mounted on the arm time such a deployment in order to minimize the impact forces to be experienced at contact. In the third phase a rigid connection between the asteroid and the end-effector is established. In the last phase, the position of the base of the spacecraft relative to contact point is autonomously adjusted to minimize the energy consumption required to despin and stabilize the joint system by maximizing the torque’s arm. This procedure relaxes the need for accurate estimates of the inertial properties of the asteroid. The concept proposed is illustrated by simulating the capture of a rigid asteroid subject to a one axis rotation with no precession or nutation.

Verification of a special inertial measurement unit using a Quadrotor aircraft

Purpose – Calibration and 6-DOF test of a unique inertial measurement unit (IMU) using a Quadrotor aircraft. The purpose of this paper is to discuss the above issue. Design/methodology/approach – An IMU with the special capability of measuring the angular acceleration was developed and tested. A Quadrotor aircraft is used as 6-DOF test platform. Kinematics modeling of the Quadrotor was used in the determination of the Euler angles, while Dynamics modeling aided in the design the closed loop controller. For safety, the flight test was performed on a 6-DOF constrained reduced-gravity test stand. Findings – The developed IMU is suitable for measuring states and its time derivatives of mini UAVs. Not only that but also a simple control algorithm can be integrated in the same processing unit (a 32 microcontroller in this case). Originality/value – The tested IMU as well as the safety constrained test techniques are unique.