About

The Space Engineering Research Center (SERC) is a joint venture between two components of the University of Southern California —the Information Sciences Institute and the Department of Astronautical Engineering. The Center is dedicated to disruptive space engineering, research, and education–including hands-on build, test and flight demonstrations of spacecraft and satellites.

Research topics

• Computer Science
• Aerospace engineering
• Artificial Intelligence
• Engineering
• Machine Learning
• Business
• Algorithm
• Physics
• Systems engineering
• Operations research

Selected publications

• Demonstration of Metrics for Self-Healing and Self-Reconfiguration in a Monolithic and Segmented Robotic System

  2026-01-08

  articleSenior author

  Methods for self-healing have been extensively studied in the context of modular, segmented systems. However, concepts of self-healing are not limited only to systems in which the constituent modules are identical. In order to quantify and study the characteristics of a system equipped with the ability to self-heal, a generalized system of metrics is required to take objective measurement of self-healing outcomes for both monolithic and segmented morphologies. This is required to generalize the concepts of self-healing across robotic systems with generic morphologies, and to provide inputs to risk analyses done in a trade study for the addition of self-healing capabilities. This paper proposes a general set of metrics to analyze healing capabilities and demonstrates the use of select metrics on two robotic systems of varying morphologies. Each system will demonstrate healing from an imposed failure, after which the resultant healed performance will be analyzed under the proposed framework.

  PublisherDOI

• Dispelling the Myth of the Universal Interface: Results and Inferences From an Evaluation of In-space Connectable Interfaces

  2026-01-08

  article1st authorCorresponding

  Arkisys is developing a long-duration orbital platform – The Port Module – capable of accepting arrivals of new payloads, technologies, components, etc. As a unique platform the method of connectivity becomes the critical attribute that requires a deep understanding and operational experience base, which translates into what is a “connectable interface”. On the path to understand and select connectable interfaces that allow for commercial operations, in early 2021 Arkisys conducted a survey of legacy, new, and currently-in-development physical interfaces for in-space connection. This survey included 33 devices, focusing on aggregate-able post-post launch connection, and the quality of that connection. Interfaces were evaluated on technical, safety, and business viability attributes. In 2025, Arkisys conducted an updated evaluation to reflect the current landscape of physical interfaces for use in in-space operations, including In-space Servicing, Assembly, and Manufacturing (ISAM). This study evaluated 38 interfaces across four functional areas, dropping several interfaces from the 2020 study which didn’t make it to market, and adding several emerging interfaces from 2020-2025. Information was collected regarding these interfaces through publicly available channels. Neither survey considered human-rated docking interfaces, as these have been studied in detail by NASA for use on the International Space Station and Lunar Gateway, which we determined may be not practical for robotic ISAM use at their current size and mass/volume. If, in the future, scaled down variants of these interfaces are available, the results of this study will be updated to include these new interfaces. This latest study was meant to combine the results of the two studies and use the data to determine commonalities and functional similarities in the designs, as well as identify suitable real-world use cases for each of these interfaces. From the results of the study and the analysis performed, first analysis pointed away from a single common interface for ISAM (i.e. a “USB for Space”) and instead towards a subset of interfaces that were optimized for niche tasks. This paper will dive into the data and analysis to discuss.

  PublisherDOI

• Machine Learning Approaches to Position Estimation in SMA-Actuated Soft Robotic Systems

  2026-01-08

  articleSenior author

  Shape Memory Alloy (SMA) actuated soft robotic systems offer advantages in adaptability and compliance but present significant challenges for modeling and control due to nonlinear and time-dependent behavior. This work presents a data-driven approach for estimating the three-dimensional position of SMA-actuated soft robotic limbs using recurrent neural networks (RNNs). A multi-limbed soft robotic platform equipped with embedded temperature, current, and inertial measurement sensors was developed, with motion-capture data providing ground-truth position labels. A randomized actuation algorithm enabled varied and continuous data collection across independent and coupled limb activations. Advanced Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) architectures with residual connections and layer normalization were trained and evaluated using synchronized multimodal sensor data across multiple lookback window configurations. The best-performing GRU model (240- timestep lookback) achieved a root mean squared error (RMSE) of 0.63 units, outperforming the best LSTM model (0.85 units RMSE). The GRU exhibited lower mean absolute error (0.19 vs. 0.35 units) and a reduced 95th percentile error (0.89 vs. 1.66 units), indicating improved robustness. Per-marker analysis indicated that the GRU more effectively modeled complex limb motion patterns, particularly for markers experiencing greater displacement and motion complexity. The results demonstrate the feasibility of applying RNN-based models for accurate, sensor-driven position estimation in SMA-actuated soft robots, providing a foundation for future closed-loop control and onboard state estimation implementations.

  PublisherDOI

• Sensor Contribution Analysis for Multimodal Estimation in Soft Robotic Actuators

  2026-01-08

  articleSenior author

  Soft robotic systems present substantial challenges for state estimation due to their nonlinear actuation dynamics, material hysteresis, and underactuated geometries. These challenges are especially pronounced in Shape Memory Alloy (SMA) driven platforms, where mechanical motion is coupled with thermoelectric behavior that varies over time. Learning-based models such as Gated Recurrent Units (GRUs) have demonstrated strong predictive performance for soft robotic state estimation, but the role of individual sensor inputs is often unclear. This work investigates sensor contribution and redundancy within a GRU-based forward model trained to estimate the three-dimensional position of an SMA-actuated soft robotic limb. A structured ablation framework is employed to train separate models with specific sensor modalities omitted from the input data, including current, infrared temperature, and inertial measurements, enabling quantitative assessment of each modality’s impact on prediction accuracy. Results from 31 ablation configurations indicate that reduced sensor sets have the potential to outperform the full multimodal suite. Excluding the IR and PWM sensors improved the root mean squared error (RMSE) by 0.5% relative to the baseline full-sensor configuration, while reducing input dimensionality by 44%. Conversely, removing IMU sensors, particularly accelerometers, caused the greatest performance degradation, indicating their unique importance for capturing dynamic motion. Models trained with only accelerometer inputs showed 11.6% degradation compared to baseline, demonstrating that while accelerometer data is critical, it benefits from complementary sensor information. These results show that selective use of sensors can improve both estimation accuracy and computational efficiency. The proposed ablation-based interpretability framework provides a generalizable method for identifying critical sensing modalities and optimizing embedded architectures for future embedded and autonomous soft robotic systems.

  PublisherDOI

• Design of Preloaded SMA-Driven Soft Robotic Limb With Onboard Thermal Sensing

  2026-01-08

  articleSenior author

  This paper presents a mechanical and sensing framework for enhancing the performance of Shape Memory Alloy (SMA) actuated soft robotic systems. A novel double-helix exoskeleton structure is presented, experimentally validated to provide an optimal SMA preload range of 2.05 N to 3.63 N. This consistent tension significantly reduces actuation uncertainties and maximizes stroke. Another contribution is an onboard thermal sensing system capable of providing real-time temperature feedback without compromising the limb's compliance. The onboard thermal sensing system compares a single non-contact IR sensor against multiple thermocouples, demonstrating that IR sensing provides higher onboard accuracy and consistency while being significantly simpler to integrate and immune to compliance/detachment issues. These contributions lay the groundwork for more predictable and controllable SMA-based actuators, supporting the future implementation of data-driven models to improve locomotion control and state estimation in soft robotic platforms.

  PublisherDOI

• Status of the Third Miniature Sensor Technology Integration Satellite Mission

  Digital Commons - USU (Utah State University) · 2025-08-28

  articleOpen access1st authorCorresponding

  The MSTI-3 satellite is the third in a series established to test, in realistic scenarios, miniature spacecraft and sensor technologies for missile detection and tracking on low-cost, low-earth orbit technology demonstration satellites. Cooperative demonstrations are planned to combine MSTI-provided target track file information, with interceptor technology tests, to fully demonstrate technologies associated with theater missile defense (TMO) targeting. The program is sponsored by the Ballistic Missile Defense Organization (BMDO) and executed by a government/industry team led by the Air Force Phillips Laboratories' Space Experiments Directorate operating location at Edwards Air Force Base.

  PublisherOA PDFDOI

• Optimizing Element & System Compliance of Robotic, Gecko Adhesion-Based Grippers to the Unknown Geometries of Space Debris Targets

  2024-01-01 · 1 citations

  articleSenior author
  PublisherDOI

• STARFISH: A Soft Bioinspired Robot Actuated by Shape-Memory Alloy Wires for On-Orbit Mapping and Manipulation

  2024-01-01

  articleSenior author
  PublisherDOI

• A Levitated Displacement Interferometer Thrust Stand for Characterization of Propellantless Propulsion Concepts

  2024-01-01

  articleSenior author
  PublisherDOI

• Multi-Source Sensor Fusion: Challenges and Opportunities for the Future of Space Operations

  2024-01-01 · 2 citations

  articleSenior author
  PublisherDOI

Frequent coauthors

• Rahul Rughani

  18 shared

• Peter Will

  University of Siegen

  15 shared

• Sriram Narayanan

  9 shared

• Everett Maness

  University of Southern California

  9 shared

• Amrita Singh

  8 shared

• Oswin Almeida

  University of Southern California

  8 shared

• R.A. Rogers

  University of Southern California

  8 shared

• Tim Barrett

  8 shared

Labs

• Space Engineering Research CenterPI

Awards & honors

• Viterbi Use Inspired Research Award (2023)
• Viterbi Engineering USC Mentoring Award for Astronautics and…
• National AIAA Space Systems Award (2003)
• National AIAA Space Systems Award (2007)