Active defense strategies have been developed to counter the threat of deepfake technology. However, a primary challenge is their lack of persistence, as their effectiveness is often short-lived. Attackers can bypass these defenses by simply collecting protected samples and retraining their models. This means that static defenses inevitably fail when attackers retrain their models, which severely limits practical use. We argue that an effective defense not only distorts forged content but also blocks the model’s ability to adapt, which occurs when attackers retrain their models on protected images. To achieve this, we propose an innovative Two-Stage Defense Framework (TSDF). Benefiting from the intensity separation mechanism designed in this paper, the framework uses dual-function adversarial perturbations to perform two roles. First, it can directly distort the forged results. Second, it acts as a poisoning vehicle that disrupts the data preparation process essential for an attacker’s retraining pipeline. By poisoning the data source, TSDF aims to prevent the attacker’s model from adapting to the defensive perturbations, thus ensuring the defense remains effective long-term. Comprehensive experiments show that the performance of traditional interruption methods degrades sharply when these methods are subjected to adversarial retraining. However, our framework shows a strong dual defense capability, which can improve the persistence of active defense. Our code will be available at https://github.com/vpsg-research/TSDF.

Robot localization is an inverse problem of finding a robot's pose using a map and sensor measurements. In recent years, Invertible Neural Networks (INN s) have successfully solved ambiguous inverse problems in various fields. This paper proposes a framework that approaches the localization problem with INN. We design a network that provides implicit map representation in the forward path and localization in the inverse path. By sampling the latent space in evaluation, Local_INN outputs robot poses with covariance, which can be used to estimate the uncertainty. We show that the localization performance of Local_INN is on par with current methods with much lower latency. We show detailed 2D and 3D map reconstruction from Local_INN using poses exterior to the training set. We also provide a global localization algorithm using Local_INN to tackle the kidnapping problem.

Safety-critical Autonomous Systems require trustworthy and transparent decision-making process to be deployable in the real world. The advancement of Machine Learning introduces high performance but largely through black-box algorithms. In this position paper, we focus the discussion of explainability specifically with Autonomous Vehicles (AVs). As a safety-critical system, AVs provide the unique opportunity to utilize cutting-edge Machine Learning techniques while requiring transparency in decision making. Interpretability in every action the AV takes becomes crucial in post-hoc analysis where blame assignment might be necessary. In this paper, we provide positioning on how researchers could consider incorporating explainability and interpretability into design and optimization of separate Autonomous Vehicle modules including Perception, Planning, and Control.

High-performance autonomy often must operate at the boundaries of safety. When external agents are present in a system, the process of ensuring safety without sacrificing performance becomes extremely difficult. In this paper we present an approach to stress test such systems based on the rapidly exploring random tree (RRT) algorithm.We propose to find faults in such systems through adversarial agent perturbations, where the behaviors of other agents in an otherwise fixed scenario are modified. This creates a large search space of possibilities, which we explore both randomly and with a focused strategy that runs RRT in a bounded projection of the observable states that we call the objective space. The approach is applied to generate tests for evaluating overtaking logic and path planning algorithms in autonomous racing, where the vehicles are driving at high speed in an adversarial environment. We evaluate several autonomous racing path planners, finding numerous collisions during overtake maneuvers in all planners. The focused RRT search finds several times more crashes than the random strategy, and, for certain planners, tens to hundreds of times more crashes in the second half of the track.

The 3rd Japan Automotive AI Challenge was an international online autonomous racing challenge where 164 teams competed in December 2021. This paper outlines the winning strategy to this competition, and the advantages and challenges of using the Autoware.Auto open source autonomous driving platform for multi-agent racing. Our winning approach includes a lane-switching opponent overtaking strategy, a global raceline optimization, and the integration of various tools from Autoware.Auto including a Model-Predictive Controller. We describe the use of perception, planning and control modules for high-speed racing applications and provide experience-based insights on working with Autoware.Auto. While our approach is a rule-based strategy that is suitable for non-interactive opponents, it provides a good reference and benchmark for learning-enabled approaches.

Continuous change detection (CCD) is one of the most famous algorithms in remote sensing time series change detection, making any improvements on it are of great significance for the practice of monitoring land cover dynamics. Inspired by the directionality of CCD, a bidirectional CCD (Bi-CCD) is proposed to improve the accuracy of change detection by selecting the optimal result from the results of CCD running from both the forward and backward directions of time series. Three criteria including root mean square error (RMSE), Akaike information criterion (AIC), and Bayesian information criterion (BIC) were adopted to define the “optimal” in this study, and their effects under three common parameter settings were evaluated in a simulation dataset. The quantitative results show that Bi-CCD based on different result selection criteria can indeed improve the accuracy of change detection in terms of omission error and commission error. In general, Bi-CCD based on RMSE achieved the lowest omission error, while Bi-CCD based on BIC obtained the lowest commission error. Compared with the unidirectional CCD, Bi-CCD reduces the omission error and commission error by at most 11.19% and 15.08%, respectively. In addition, the different effects of different optimal result selection criteria on the accuracy of Bi-CCD enable Bi-CCD more flexible to handle different tasks than the standard CCD.

An effective network structure is essential for the classification of satellite image time series (SITS). Deep learning models have been widely used for SITS classification and achieved impressive performance, especially the architectures based on self-attention. However, the lack of efficient and comprehensive attention to valuable bands and time series structure hinders the performance to some extent. To address this problem, an end-to-end attention-aware dynamic self-aggregation network (ADSN) is proposed for SITS classification in this work, which combines two main parts: spectral focusing and spectral–temporal feature learning. The core components of ADSN are the channel attention module and dynamic self-aggregation block. Specifically, informative bands in the SITS flowing through the channel attention module can adaptively get a high weight to increase their contributions, while the attentions of some low-efficiency bands are weakened. Besides, the dynamic self-aggregation block, which integrates multiscale dynamic convolution and improved multihead attention in parallel, can simultaneously capture long- and short-distance sequence structures and position relationships to better represent temporal information. Compared with random forest (RF) and seven deep learning algorithms, the proposed model effectively learns spectral and temporal features, and the experimental results confirm that ADSN has achieved superior classification accuracy and generalization ability on two SITS datasets with extremely unbalanced samples.

Balancing performance and safety is crucial to deploying autonomous vehicles in multi-agent environments. In particular, autonomous racing is a domain that penalizes safe but conservative policies, highlighting the need for robust, adaptive strategies. Current approaches either make simplifying assumptions about other agents or lack robust mechanisms for online adaptation. This work makes algorithmic contributions to both challenges. First, to generate a realistic, diverse set of opponents, we develop a novel method for self-play based on replica-exchange Markov chain Monte Carlo. Second, we propose a distributionally robust bandit optimization procedure that adaptively adjusts risk aversion relative to uncertainty in beliefs about opponents’ behaviors. We rigorously quantify the tradeoffs in performance and robustness when approximating these computations in real-time motion-planning, and we demonstrate our methods experimentally on autonomous vehicles that achieve scaled speeds comparable to Formula One racecars.

TUNERCAR is a toolchain that jointly optimizes racing strategy, planning methods, control algorithms, and vehicle parameters for an autonomous racecar. In this paper, we detail the target hardware, software, simulators, and systems infrastructure for this toolchain. Our methodology employs a parallel implementation of CMA-ES which enables simulations to proceed 6 times faster than real-world rollouts. We show our approach can reduce the lap times in autonomous racing, given a fixed computational budget. For all tested tracks, our method provides the lowest lap time, and relative improvements in lap time between 7-21%. We demonstrate improvements over a naive random search method with equivalent computational budget of over 15 seconds/lap, and improvements over expert solutions of over 2 seconds/lap. We further compare the performance of our method against hand-tuned solutions submitted by over 30 international teams, comprised of graduate students working in the field of autonomous vehicles. Finally, we discuss the effectiveness of utilizing an online planning mechanism to reduce the reality gap between our simulation and actual tests.

In this letter, multisource earth observation (EO) data sets, including multitemporal Landsat-8, digital surface model, and spatial information, were integrated for land-cover classification by random forest (RF) and support vector machines (SVMs). We demonstrated in this letter that both RF and SVM are useful tools for classification of land cover in the local climate zones featured with highly heterogeneous landscape. Classification of land cover by RF was with an overall accuracy (OA) of 86.2%, while the OA was 85.5% for SVM. However, we found that RF was more stable than SVM for multisource EO data in classifying land cover without normalizing different feature data sets. Experiments showed that the thermal features were more important than temporal and spatial ones in discriminating impervious objects, while the temporal and spatial features were generally better than thermal ones in separating the distinct vegetation categories. Another finding was that our experiments indicated that spectral features were the most important in classification of land cover, followed by temporal, thermal, and spatial features, respectively. As to the spectral features, red channels were the most important, followed by short-wave infrared, near-infrared, and green channels. Thus, it could be concluded that the combination of spectral, thermal, spatial, and temporal information would be an optimal approach to increase the OA of land-cover classification in the zones featured with highly heterogeneous landscape.

We present an approach that allows the Georgia Tech Miniature Autonomous Blimp (GT-MAB) to detect and follow a human. This accomplishment is the first Human Robot Interaction (HRI) demonstration between an uninstrumented human and a robotic blimp. GT-MAB is an ideal platform for HRI missions because it is safe to humans and can support sufficient flight time for HRI experiments. However, due to complex aerodynamic influence on the blimp, the human following task for GT-MAB with a single on-board camera is a challenging problem. We integrate Haar face detector and KLT feature tracker to achieve robust human tracking. After a human face is detected in the real-time video stream, we estimated the 3D positions of the human with respect to GT-MAB. Visionbased PID controllers are designed based on estimated relative position and the motion primitives of GT-MAB such that it can achieve stable and continuous human following behavior. Experimental results are presented to demonstrate the human following capability on GT-MAB.