In this paper, we present a novel fog-aware object detection network called FogGuard, designed to address the challenges posed by foggy weather conditions. Autonomous driving systems heavily rely on accurate object detection algorithms, but adverse weather conditions can significantly impact the reliability of deep neural networks (DNNs). Existing approaches fall into two main categories, 1) image enhancement such as IA-YOLO 2) domain adaptation based approaches. Image enhancement based techniques attempt to generate fog-free image. However, retrieving a fogless image from a foggy image is a much harder problem than detecting objects in a foggy image. Domain-adaptation based approaches, on the other hand, do not make use of labelled datasets in the target domain. Both categories of approaches are attempting to solve a harder version of the problem. Our approach builds over fine-tuning on the Our framework is specifically designed to compensate for foggy conditions present in the scene, ensuring robust performance even. We adopt YOLOv3 as the baseline object detection algorithm and introduce a novel Teacher-Student Perceptual loss, to high accuracy object detection in foggy images. Through extensive evaluations on common datasets such as PASCAL VOC and RTTS, we demonstrate the improvement in performance achieved by our network. We demonstrate that FogGuard achieves 69.43\% mAP, as compared to 57.78\% for YOLOv3 on the RTTS dataset. Furthermore, we show that while our training method increases time complexity, it does not introduce any additional overhead during inference compared to the regular YOLO network.
This paper considers safe control synthesis for dynamical systems in the presence of uncertainty in the dynamics model and the safety constraints that the system must satisfy. Our approach captures probabilistic and worst-case model errors and their effect on control Lyapunov function (CLF) and control barrier function (CBF) constraints in the control-synthesis optimization problem. We show that both the probabilistic and robust formulations lead to second-order cone programs (SOCPs), enabling safe and stable control synthesis that can be performed efficiently online. We evaluate our approach in PyBullet simulations of an autonomous robot navigating in unknown environments and compare the performance with a baseline CLF-CBF quadratic programming approach.
This paper focuses on inverse reinforcement learning for autonomous navigation using distance and semantic category observations. The objective is to infer a cost function that explains demonstrated behavior while relying only on the expert's observations and state-control trajectory. We develop a map encoder, that infers semantic category probabilities from the observation sequence, and a cost encoder, defined as a deep neural network over the semantic features. Since the expert cost is not directly observable, the model parameters can only be optimized by differentiating the error between demonstrated controls and a control policy computed from the cost estimate. We propose a new model of expert behavior that enables error minimization using a closed-form subgradient computed only over a subset of promising states via a motion planning algorithm. Our approach allows generalizing the learned behavior to new environments with new spatial configurations of the semantic categories. We analyze the different components of our model in a minigrid environment. We also demonstrate that our approach learns to follow traffic rules in the autonomous driving CARLA simulator by relying on semantic observations of buildings, sidewalks, and road lanes.
This paper focuses on learning a model of system dynamics online while satisfying safety constraints. Our objective is to avoid offline system identification or hand-specified models and allow a system to safely and autonomously estimate and adapt its own model during operation. Given streaming observations of the system state, we use Bayesian learning to obtain a distribution over the system dynamics. Specifically, we use a matrix variate Gaussian process (MVGP) regression approach with efficient covariance factorization to learn the drift and input gain terms of a nonlinear control-affine system. The MVGP distribution is then used to optimize the system behavior and ensure safety with high probability, by specifying control Lyapunov function (CLF) and control barrier function (CBF) chance constraints. We show that a safe control policy can be synthesized for systems with arbitrary relative degree and probabilistic CLF-CBF constraints by solving a second order cone program (SOCP). Finally, we extend our design to a self-triggering formulation, adaptively determining the time at which a new control input needs to be applied in order to guarantee safety.
This paper focuses on inverse reinforcement learning (IRL) for autonomous robot navigation using semantic observations. The objective is to infer a cost function that explains demonstrated behavior while relying only on the expert's observations and state-control trajectory. We develop a map encoder, which infers semantic class probabilities from the observation sequence, and a cost encoder, defined as deep neural network over the semantic features. Since the expert cost is not directly observable, the representation parameters can only be optimized by differentiating the error between demonstrated controls and a control policy computed from the cost estimate. The error is optimized using a closed-form subgradient computed only over a subset of promising states via a motion planning algorithm. We show that our approach learns to follow traffic rules in the autonomous driving CARLA simulator by relying on semantic observations of cars, sidewalks and road lanes.
This paper focuses on inverse reinforcement learning (IRL) to enable safe and efficient autonomous navigation in unknown partially observable environments. The objective is to infer a cost function that explains expert-demonstrated navigation behavior while relying only on the observations and state-control trajectory used by the expert. We develop a cost function representation composed of two parts: a probabilistic occupancy encoder, with recurrent dependence on the observation sequence, and a cost encoder, defined over the occupancy features. The representation parameters are optimized by differentiating the error between demonstrated controls and a control policy computed from the cost encoder. Such differentiation is typically computed by dynamic programming through the value function over the whole state space. We observe that this is inefficient in large partially observable environments because most states are unexplored. Instead, we rely on a closed-form subgradient of the cost-to-go obtained only over a subset of promising states via an efficient motion-planning algorithm such as A* or RRT. Our experiments show that our model exceeds the accuracy of baseline IRL algorithms in robot navigation tasks, while substantially improving the efficiency of training and test-time inference.
This paper focuses on learning a model of system dynamics online while satisfying safety constraints. Our motivation is to avoid offline system identification or hand-specified dynamics models and allow a system to safely and autonomously estimate and adapt its own model during online operation. Given streaming observations of the system state, we use Bayesian learning to obtain a distribution over the system dynamics. In turn, the distribution is used to optimize the system behavior and ensure safety with high probability, by specifying a chance constraint over a control barrier function.
Learning from Demonstrations (LfD) via Behavior Cloning (BC) works well on multiple complex tasks. However, a limitation of the typical LfD approach is that it requires expert demonstrations for all scenarios, including those in which the algorithm is already well-trained. The recently proposed Learning from Interventions (LfI) overcomes this limitation by using an expert overseer. The expert overseer only intervenes when it suspects that an unsafe action is about to be taken. Although LfI significantly improves over LfD, the state-of-the-art LfI fails to account for delay caused by the expert's reaction time and only learns short-term behavior. We address these limitations by 1) interpolating the expert's interventions back in time, and 2) by splitting the policy into two hierarchical levels, one that generates sub-goals for the future and another that generates actions to reach those desired sub-goals. This sub-goal prediction forces the algorithm to learn long-term behavior while also being robust to the expert's reaction time. Our experiments show that LfI using sub-goals in a hierarchical policy framework trains faster and achieves better asymptotic performance than typical LfD.
The navigation problem is classically approached in two steps: an exploration step, where map-information about the environment is gathered; and an exploitation step, where this information is used to navigate efficiently. Deep reinforcement learning (DRL) algorithms, alternatively, approach the problem of navigation in an end-to-end fashion. Inspired by the classical approach, we ask whether DRL algorithms are able to inherently explore, gather and exploit map-information over the course of navigation. We build upon Mirowski et al. [2017] work and introduce a systematic suite of experiments that vary three parameters: the agent's starting location, the agent's target location, and the maze structure. We choose evaluation metrics that explicitly measure the algorithm's ability to gather and exploit map-information. Our experiments show that when trained and tested on the same maps, the algorithm successfully gathers and exploits map-information. However, when trained and tested on different sets of maps, the algorithm fails to transfer the ability to gather and exploit map-information to unseen maps. Furthermore, we find that when the goal location is randomized and the map is kept static, the algorithm is able to gather and exploit map-information but the exploitation is far from optimal. We open-source our experimental suite in the hopes that it serves as a framework for the comparison of future algorithms and leads to the discovery of robust alternatives to classical navigation methods.
Consider mutli-goal tasks that involve static environments and dynamic goals. Examples of such tasks, such as goal-directed navigation and pick-and-place in robotics, abound. Two types of Reinforcement Learning (RL) algorithms are used for such tasks: model-free or model-based. Each of these approaches has limitations. Model-free RL struggles to transfer learned information when the goal location changes, but achieves high asymptotic accuracy in single goal tasks. Model-based RL can transfer learned information to new goal locations by retaining the explicitly learned state-dynamics, but is limited by the fact that small errors in modelling these dynamics accumulate over long-term planning. In this work, we improve upon the limitations of model-free RL in multi-goal domains. We do this by adapting the Floyd-Warshall algorithm for RL and call the adaptation Floyd-Warshall RL (FWRL). The proposed algorithm learns a goal-conditioned action-value function by constraining the value of the optimal path between any two states to be greater than or equal to the value of paths via intermediary states. Experimentally, we show that FWRL is more sample-efficient and learns higher reward strategies in multi-goal tasks as compared to Q-learning, model-based RL and other relevant baselines in a tabular domain.