A5006271370- Robotic Path Planning Algorithms
- Robotics and Sensor-Based Localization
- Maritime Navigation and Safety
- Autonomous Vehicle Technology and Safety
- Target Tracking and Data Fusion in Sensor Networks
- Fault Detection and Control Systems
- Housing Market and Economics
- Advanced Control Systems Optimization
- Modular Robots and Swarm Intelligence
- Underwater Vehicles and Communication Systems
- Vehicle Dynamics and Control Systems
- Traffic Prediction and Management Techniques
- Real-time simulation and control systems
- Industrial Technology and Control Systems
- Infrared Target Detection Methodologies
- Aerospace Engineering and Control Systems
- Domain Adaptation and Few-Shot Learning
- Reinforcement Learning in Robotics
- Adaptive Control of Nonlinear Systems
- Urban Planning and Valuation
- Simulation Techniques and Applications
- Complex Systems and Decision Making
- Iterative Learning Control Systems
- Robotic Locomotion and Control
- Robotics and Automated Systems
Jet Propulsion Laboratory
2023-2024
California Institute of Technology
2020-2024
Synchronized control of propellers and legs enables a bipedal robot to fly, slackline, skateboard.
Sampling-based model-predictive controllers have become a powerful optimization tool for planning and control problems in various challenging environments.In this paper, we show how the default choice of uncorrelated Gaussian distributions can be improved upon with use colored noise distribution.Our distribution allows emphasis on low frequency signals, which result smoother more exploratory samples.We frequency-based sampling Model Predictive Path Integral (MPPI) both hardware simulation...
Fixed-wing vertical take-off and landing (VTOL) aircraft pose a unique control challenge that stems from complex aerodynamic interactions between wings rotors. Thus, accurate estimation of external forces is indispensable for achieving high performance flight. In this paper, we present composite adaptive nonlinear tracking controller fixed-wing VTOL. The method employs online adaptation linear force models, generates wing rotor in real-time based on information three-dimensional airflow...
Autonomous navigation at high speeds in off-road environments necessitates robots to comprehensively understand their surroundings using onboard sensing only. The extreme conditions posed by the setting can cause degraded camera image quality due poor lighting and motion blur, as well limited sparse geometric information available from LiDAR when driving speeds. In this work, we present RoadRunner, a novel framework capable of predicting terrain traversability an elevation map directly...
Understanding terrain topology at long-range is crucial for the success of off-road robotic missions, especially when navigating high-speeds. LiDAR sensors, which are currently heavily relied upon geometric mapping, provide sparse measurements mapping greater distances. To address this challenge, we present a novel learning-based approach capable predicting elevation maps longrange using only onboard egocentric images in real-time. Our proposed method comprised three main elements. First,...
Reliable offroad autonomy requires low-latency, high-accuracy state estimates of pose as well velocity, which remain viable throughout environments with sub-optimal operating conditions for the utilized perception modalities. As estimation remains a single point failure system in majority aspiring autonomous systems, failing to address environmental degradation sensors could potentially experience given conditions, can be mission-critical shortcoming. In this work, method integration radar...
Modeling dynamics is often the first step to making a vehicle autonomous. While on-road autonomous vehicles have been extensively studied, off-road pose many challenging modeling problems. An encounters highly complex and difficult-to-model terrain/vehicle interactions, as well having of its own. These complexities can create challenges for effective high-speed control planning. In this paper, we introduce framework multistep prediction that explicitly handles accumulation error remains...
View Video Presentation: https://doi.org/10.2514/6.2021-1514.vid As electric aircraft take to the skies, it is becoming common speculate on potential niches in which apply technology. One promising role for these vehicles medical domain, a currently filled by helicopters. A Vertical Take-Off and Landing (VTOL) with autonomous capabilities could avoid obstacles transport injured patients receive critical care more quickly than land-based options. This paper presents design of novel,...
Reliable offroad autonomy requires low-latency, high-accuracy state estimates of pose as well velocity, which remain viable throughout environments with sub-optimal operating conditions for the utilized perception modalities. As estimation remains a single point failure system in majority aspiring autonomous systems, failing to address environmental degradation sensors could potentially experience given conditions, can be mission-critical shortcoming. In this work, method integration radar...
Understanding terrain topology at long-range is crucial for the success of off-road robotic missions, especially when navigating high-speeds. LiDAR sensors, which are currently heavily relied upon geometric mapping, provide sparse measurements mapping greater distances. To address this challenge, we present a novel learning-based approach capable predicting elevation maps using only onboard egocentric images in real-time. Our proposed method comprised three main elements. First,...
Sampling-based model-predictive controllers have become a powerful optimization tool for planning and control problems in various challenging environments. In this paper, we show how the default choice of uncorrelated Gaussian distributions can be improved upon with use colored noise distribution. Our distribution allows emphasis on low frequency signals, which result smoother more exploratory samples. We frequency-based sampling Model Predictive Path Integral (MPPI) both hardware simulation...
Autonomous robot navigation in off-road environments requires a comprehensive understanding of the terrain geometry and traversability. The degraded perceptual conditions sparse geometric information at longer ranges make problem challenging especially when driving high speeds. Furthermore, sensing-to-mapping latency look-ahead map range can limit maximum speed vehicle. Building on top recent work RoadRunner, this work, we address challenge long-range (100 m) traversability estimation. Our...
Enabling robot autonomy in complex environments for mission critical application requires robust state estimation. Particularly under conditions where the exteroceptive sensors, which navigation depends on, can be degraded by environmental challenges thus, leading to failure. It is precisely such potential FMCW radar sensors highlighted: as a complementary sensing modality with direct velocity measuring capabilities. In this work we integrate radial speed measurements from sensor, using...
Off-road environments pose significant perception challenges for high-speed autonomous navigation due to unstructured terrain, degraded sensing conditions, and domain-shifts among biomes. Learning semantic information across these conditions biomes can be challenging when a large amount of ground truth data is required. In this work, we propose an approach that leverages pre-trained Vision Transformer (ViT) with fine-tuning on small (<500 images), sparse coarsely labeled (<30% pixels)...
Rapid autonomous traversal of unstructured terrain is essential for scenarios such as disaster response, search and rescue, or planetary exploration. As a vehicle navigates at the limit its capabilities over extreme terrain, dynamics can change suddenly dramatically. For example, high-speed varying affect parameters traction, tire slip, rolling resistance. To achieve effective planning in environments, it crucial to have model that accurately anticipate these conditions. In this work, we...
We propose TRADE for robust tracking and 3D localization of a moving target in complex environments, from UAVs equipped with single camera. Ultimately enables 3d-aware following. Tracking-by-detection approaches are vulnerable to switching, especially between similar objects. Thus, predicts incorporates the trajectory select right tracker's response map. Unlike static depth estimation camera is an ill-posed problem. Therefore we novel method ground targets on terrain. It reasons about scene...
Modeling dynamics is often the first step to making a vehicle autonomous. While on-road autonomous vehicles have been extensively studied, off-road pose many challenging modeling problems. An encounters highly complex and difficult-to-model terrain/vehicle interactions, as well having of its own. These complexities can create challenges for effective high-speed control planning. In this paper, we introduce framework multistep prediction that explicitly handles accumulation error remains...
PARSEC (Payload Anchoring Robotic System for the Exploration of Cliffs) is an autonomy-equipped aerial manipulator that can deploy self-anchoring payloads on rocky vertical surfaces. It consists a hexacopter and two Degrees Freedom (2 DoF) mass balancing manipulator, which autonomously payload from its custom end-effector. The anchors itself via actuated microspine gripper. Payload sensor data wirelessly transmitted to primary vehicle during after deployment. A novel state machine controls...
Holistic real estate strategies have to fulfill different interests and demands. Not only do qualitative quantitative goals be reconciled, but also conflicting resolved. This paper investigates how create a rating that captures the dynamic developments of over its life cycle. A proprietary model is presented based on dynamically evolving utility analysis takes into account both changes in condition changing requirements user groups. for therefore not offers possibility depicting future...