Grace Gao

Mahmoud Selim, Amr Alanwar, Shreyas Kousik et al. · gatech

Reinforcement learning (RL) is capable of sophisticated motion planning and control for robots in uncertain environments. However, state-of-the-art deep RL approaches typically lack safety guarantees, especially when the robot and environment models are unknown. To justify widespread deployment, robots must respect safety constraints without sacrificing performance. Thus, we propose a Black-box Reachability-based Safety Layer (BRSL) with three main components: (1) data-driven reachability analysis for a black-box robot model, (2) a trajectory rollout planner that predicts future actions and observations using an ensemble of neural networks trained online, and (3) a differentiable polytope collision check between the reachable set and obstacles that enables correcting unsafe actions. In simulation, BRSL outperforms other state-of-the-art safe RL methods on a Turtlebot 3, a quadrotor, a trajectory-tracking point mass, and a hexarotor in wind with an unsafe set adjacent to the area of highest reward.

Anthony Corso, Kyu-Young Kim, Shubh Gupta et al.

An important step in the design of autonomous systems is to evaluate the probability that a failure will occur. In safety-critical domains, the failure probability is extremely small so that the evaluation of a policy through Monte Carlo sampling is inefficient. Adaptive importance sampling approaches have been developed for rare event estimation but do not scale well to sequential systems with long horizons. In this work, we develop two adaptive importance sampling algorithms that can efficiently estimate the probability of rare events for sequential decision making systems. The basis for these algorithms is the minimization of the Kullback-Leibler divergence between a state-dependent proposal distribution and a target distribution over trajectories, but the resulting algorithms resemble policy gradient and value-based reinforcement learning. We apply multiple importance sampling to reduce the variance of our estimate and to address the issue of multi-modality in the optimal proposal distribution. We demonstrate our approach on a control task with both continuous and discrete actions spaces and show accuracy improvements over several baselines.

Daniel Neamati, Sriramya Bhamidipati, Grace Gao

Risk-aware urban localization with the Global Navigation Satellite System (GNSS) remains an unsolved problem with frequent misdetection of the user's street or side of the street. Significant advances in 3D map-aided GNSS use grid-based GNSS shadow matching alongside AI-driven line-of-sight (LOS) classifiers and server-based processing to improve localization accuracy, especially in the cross-street direction. Our prior work introduces a new paradigm for shadow matching that proposes set-valued localization with computationally efficient zonotope set representations. While existing literature improved accuracy and efficiency, the current state of shadow matching theory does not address the needs of risk-aware autonomous systems. We extend our prior work to propose Mosaic Zonotope Shadow Matching (MZSM) that employs a classifier-agnostic polytope mosaic architecture to provide risk-awareness and certifiable guarantees on urban positioning. We formulate a recursively expanding binary tree that refines an initial location estimate with set operations into smaller polytopes. Together, the smaller polytopes form a mosaic. We weight the tree branches with the probability that the user is in line of sight of the satellite and expand the tree with each new satellite observation. Our method yields an exact shadow matching distribution from which we guarantee uncertainty bounds on the user localization. We perform high-fidelity simulations using a 3D building map of San Francisco to validate our algorithm's risk-aware improvements. We demonstrate that MZSM provides certifiable guarantees across varied data-driven LOS classifier accuracies and yields a more precise understanding of the uncertainty over existing methods. We validate that our tree-based construction is efficient and tractable, computing a mosaic from 14 satellites in 0.63 seconds and growing quadratically in the satellite number.

Adam Dai, Shubh Gupta, Grace Gao · amazon-science

Autonomous vehicles such as the Mars rovers currently lead the vanguard of surface exploration on extraterrestrial planets and moons. In order to accelerate the pace of exploration and science objectives, it is critical to plan safe and efficient paths for these vehicles. However, current rover autonomy is limited by a lack of global maps which can be easily constructed and stored for onboard re-planning. Recently, Neural Radiance Fields (NeRFs) have been introduced as a detailed 3D scene representation which can be trained from sparse 2D images and efficiently stored. We propose to use NeRFs to construct maps for online use in autonomous navigation, and present a planning framework which leverages the NeRF map to integrate local and global information. Our approach interpolates local cost observations across global regions using kernel ridge regression over terrain features extracted from the NeRF map, allowing the rover to re-route itself around untraversable areas discovered during online operation. We validate our approach in high-fidelity simulation and demonstrate lower cost and higher percentage success rate path planning compared to various baselines.

Timothy Chen, Adam Dai, Maximilian Adang et al.

What makes a good viewpoint? The quality of the data used to learn 3D reconstructions is crucial for enabling efficient and accurate scene modeling. We study the active view selection problem and develop a principled analysis that yields a simple and interpretable criterion for selecting informative camera poses. Our key insight is that informative views can be obtained by minimizing a tractable approximation of the Fisher Information Gain, which reduces to favoring viewpoints that cover geometry that has been insufficiently observed by past cameras. This leads to a lightweight coverage-based view selection metric that avoids expensive transmittance estimation and is robust to noise and training dynamics. We call this metric COVER (Camera Optimization for View Exploration and Reconstruction). We integrate our method into the Nerfstudio framework and evaluate it on real datasets within fixed and embodied data acquisition scenarios. Across multiple datasets and radiance-field baselines, our method consistently improves reconstruction quality compared to state-of-the-art active view selection methods. Additional visualizations and our Nerfstudio package can be found at https://chengine.github.io/nbv_gym/.

Adam Dai, Asta Wu, Keidai Iiyama et al.

We present a modular, full-stack autonomy system for lunar surface navigation and mapping developed for the Lunar Autonomy Challenge. Operating in a GNSS-denied, visually challenging environment, our pipeline integrates semantic segmentation, stereo visual odometry, pose graph SLAM with loop closures, and layered planning and control. We leverage lightweight learning-based perception models for real-time segmentation and feature tracking and use a factor-graph backend to maintain globally consistent localization. High-level waypoint planning is designed to promote mapping coverage while encouraging frequent loop closures, and local motion planning uses arc sampling with geometric obstacle checks for efficient, reactive control. We evaluate our approach in the competition's high-fidelity lunar simulator, demonstrating centimeter-level localization accuracy, high-fidelity map generation, and strong repeatability across random seeds and rock distributions. Our solution achieved first place in the final competition evaluation.

Keidai Iiyama, Daniel Neamati, Grace Gao

To address the need for robust positioning, navigation, and timing services in lunar environments, this paper proposes a novel onboard clock phase jump detection framework for satellite constellations using range measurements obtained from dual one-way inter-satellite links. Our approach leverages vertex redundantly rigid graphs to detect faults without relying on prior knowledge of satellite positions or clock biases, providing flexibility for lunar satellite networks with diverse satellite types and operators. We model satellite constellations as graphs, where satellites are vertices and inter-satellite links are edges. The proposed algorithm detects and identifies satellites with clock jumps by monitoring the singular values of the geometric-centered Euclidean distance matrix (GCEDM) of 5-clique sub-graphs. The proposed method is validated through simulations of a GPS constellation and a notional constellation around the Moon, demonstrating its effectiveness in various configurations.

Ashwin V. Kanhere, Shubh Gupta, Akshay Shetty et al.

Deep Neural Networks (DNNs) are a promising tool for Global Navigation Satellite System (GNSS) positioning in the presence of multipath and non-line-of-sight errors, owing to their ability to model complex errors using data. However, developing a DNN for GNSS positioning presents various challenges, such as 1) poor numerical conditioning caused by large variations in measurements and position values across the globe, 2) varying number and order within the set of measurements due to changing satellite visibility, and 3) overfitting to available data. In this work, we address the aforementioned challenges and propose an approach for GNSS positioning by applying DNN-based corrections to an initial position guess. Our DNN learns to output the position correction using the set of pseudorange residuals and satellite line-of-sight vectors as inputs. The limited variation in these input and output values improves the numerical conditioning for our DNN. We design our DNN architecture to combine information from the available GNSS measurements, which vary both in number and order, by leveraging recent advancements in set-based deep learning methods. Furthermore, we present a data augmentation strategy for reducing overfitting in the DNN by randomizing the initial position guesses. We first perform simulations and show an improvement in the initial positioning error when our DNN-based corrections are applied. After this, we demonstrate that our approach outperforms a WLS baseline on real-world data. Our implementation is available at github.com/Stanford-NavLab/deep_gnss.

Guillem Casadesus Vila, Adam Dai, Grace Gao

Navigation and mapping on the lunar surface require robust perception under challenging conditions, including poorly textured environments, high-contrast lighting, and limited computational resources. This paper presents a real-time mapping framework that integrates dense perception models with a 3D Gaussian Splatting (3DGS) representation. We first benchmark several models on synthetic datasets generated with the LuPNT simulator, selecting a stereo dense depth estimation model based on Gated Recurrent Units for its balance of speed and accuracy in depth estimation, and a convolutional neural network for its superior performance in detecting semantic segments. Using ground truth poses to decouple the local scene understanding from the global state estimation, our pipeline reconstructs a 120-meter traverse with a geometric height accuracy of approximately 3 cm, outperforming a traditional point cloud baseline without LiDAR. The resulting 3DGS map enables novel view synthesis and serves as a foundation for a full SLAM system, where its capacity for joint map and pose optimization would offer significant advantages. Our results demonstrate that combining semantic segmentation and dense depth estimation with learned map representations is an effective approach for creating detailed, large-scale maps to support future lunar surface missions.

Adam Dai, Guillem Casadesus Vila, Grace Gao

Future lunar missions will require autonomous rovers capable of traversing tens of kilometers across challenging terrain while maintaining accurate localization and producing globally consistent maps. However, the absence of global positioning systems, extreme illumination, and low-texture regolith make long-range navigation on the Moon particularly difficult, as visual-inertial odometry pipelines accumulate drift over extended traverses. To address this challenge, we present a stereo visual simultaneous localization and mapping (SLAM) system that integrates learned feature detection and matching with global constraints from digital elevation models (DEMs). Our front-end employs learning-based feature extraction and matching to achieve robustness to illumination extremes and repetitive terrain, while the back-end incorporates DEM-derived height and surface-normal factors into a pose graph, providing absolute surface constraints that mitigate long-term drift. We validate our approach using both simulated lunar traverse data generated in Unreal Engine and real Moon/Mars analog data collected from Mt. Etna. Results demonstrate that DEM anchoring consistently reduces absolute trajectory error compared to baseline SLAM methods, lowering drift in long-range navigation even in repetitive or visually aliased terrain.

Adyasha Mohanty, Grace Gao

Global Navigation Satellite Systems (GNSS)-based positioning plays a crucial role in various applications, including navigation, transportation, logistics, mapping, and emergency services. Traditional GNSS positioning methods are model-based and they utilize satellite geometry and the known properties of satellite signals. However, model-based methods have limitations in challenging environments and often lack adaptability to uncertain noise models. This paper highlights recent advances in Machine Learning (ML) and its potential to address these limitations. It covers a broad range of ML methods, including supervised learning, unsupervised learning, deep learning, and hybrid approaches. The survey provides insights into positioning applications related to GNSS such as signal analysis, anomaly detection, multi-sensor integration, prediction, and accuracy enhancement using ML. It discusses the strengths, limitations, and challenges of current ML-based approaches for GNSS positioning, providing a comprehensive overview of the field.

Sriramya Bhamidipati, Tara Mina, Grace Gao

Recently, there has been a growing interest in the use of a SmallSat platform for the future Lunar Navigation Satellite System (LNSS) to allow for cost-effectiveness and rapid deployment. However, many design choices are yet to be finalized for the SmallSat-based LNSS, including the onboard clock and the orbit type. As compared to the legacy Earth-GPS, designing an LNSS poses unique challenges: (a) restricted Size, Weight, and Power (SWaP) of the onboard clock, which limits the timing stability; (b) limited lunar ground monitoring stations, which engenders a greater preference toward stable LNSS satellite orbits. In this current work, we analyze the trade-off between different design considerations related to the onboard clock and the lunar orbit type for designing an LNSS with time-transfer from Earth-GPS. Our proposed time-transfer architecture combines the intermittently available Earth-GPS signals in a timing filter to alleviate the cost and SWaP requirements of the onboard clocks. Specifically, we conduct multiple case studies with different grades of low-SWaP clocks and various previously studied lunar orbit types. We estimate the lunar User Equivalent Range Error (UERE) metric to characterize the ranging accuracy of signals transmitted from an LNSS satellite. Using the Systems Tool Kit (STK)-based simulation setup from Analytical Graphics, Inc. (AGI), we evaluate the lunar UERE across various case studies of the LNSS design to demonstrate comparable performance as that of the legacy Earth-GPS, even while using a low-SWaP onboard clock. We further perform sensitivity analysis to investigate the variation in the lunar UERE metric across different case studies as the Earth-GPS measurement update rates are varied.

Akshay Shetty, Timmy Hussain, Grace Gao

Communication connectivity is desirable for safe and efficient operation of multi-robot systems. While decentralized algorithms for connectivity maintenance have been explored in recent literature, the majority of these works do not account for robot motion and sensing uncertainties. These uncertainties are inherent in practical robots and result in robots deviating from their desired positions which could potentially result in a loss of connectivity. In this paper we present a Decentralized Connectivity Maintenance algorithm accounting for robot motion and sensing Uncertainties (DCMU). We first propose a novel weighted graph definition for the multi-robot system that accounts for the aforementioned uncertainties along with realistic connectivity constraints such as line-of-sight connectivity and collision avoidance. Next we design a decentralized gradient-based controller for connectivity maintenance where we derive the gradients of our weighted graph edge weights required for computing the control. Finally, we perform multiple simulations to validate the connectivity maintenance performance of our DCMU algorithm under robot motion and sensing uncertainties and show an improvement compared to previous work.

Grace Gao

Navigating networked robot swarms often requires knowing where to go, sensing the environment, and path-planning based on the destination and barriers in the environment. Such a process is computationally intensive. Moreover, as the network scales up, the computational load increases quadratically, or even exponentially. Unlike these man-made systems, most biological systems scale linearly in complexity. Furthermore, the scale of a biological swarm can even enable collective intelligence. One example comes from observations of golden shiner fish. Golden shiners naturally prefer darkness and school together. Each individual golden shiner does not know where the darkness is. Neither does it sense the light gradients in the environment. However, by moving together as a school, they always end up in the shady area. We apply such collective intelligence learned from golden shiner fish to navigating robot swarms. Each individual robot's dynamic is based on the gold shiners' movement strategy---a random walk with its speed modulated by the light intensity and its direction affected by its neighbors. The theoretical analysis and simulation results show that our method 1) promises to navigate a robot swarm with little situational knowledge, 2) simplifies control and decision-making for each individual robot, 3) requires minimal or even no information exchange within the swarm, and 4) is highly distributed, adaptive, and robust.