Deep reinforcement learning has achieved great success in laser-based collision avoidance works because the laser can sense accurate depth information without too much redundant data, which can maintain the robustness of the algorithm when it is migrated from the simulation environment to the real world. However, high-cost laser devices are not only difficult to deploy for a large scale of robots but also demonstrate unsatisfactory robustness towards the complex obstacles, including irregular obstacles, e.g., tables, chairs, and shelves, as well as complex ground and special materials. In this paper, we propose a novel monocular camera-based complex obstacle avoidance framework. Particularly, we innovatively transform the captured RGB images to pseudo-laser measurements for efficient deep reinforcement learning. Compared to the traditional laser measurement captured at a certain height that only contains one-dimensional distance information away from the neighboring obstacles, our proposed pseudo-laser measurement fuses the depth and semantic information of the captured RGB image, which makes our method effective for complex obstacles. We also design a feature extraction guidance module to weight the input pseudo-laser measurement, and the agent has more reasonable attention for the current state, which is conducive to improving the accuracy and efficiency of the obstacle avoidance policy. Besides, we adaptively add the synthesized noise to the laser measurement during the training stage to decrease the sim-to-real gap and increase the robustness of our model in the real environment. Finally, the experimental results show that our framework achieves state-of-the-art performance in several virtual and real-world scenarios.

Acquiring the most representative examples via active learning (AL) can benefit many data-dependent computer vision tasks by minimizing efforts of image-level or pixel-wise annotations. In this paper, we propose a novel Collaborative Panoptic-Regional Active Learning framework (CPRAL) to address the semantic segmentation task. For a small batch of images initially sampled with pixel-wise annotations, we employ panoptic information to initially select unlabeled samples. Considering the class imbalance in the segmentation dataset, we import a Regional Gaussian Attention module (RGA) to achieve semantics-biased selection. The subset is highlighted by vote entropy and then attended by Gaussian kernels to maximize the biased regions. We also propose a Contextual Labels Extension (CLE) to boost regional annotations with contextual attention guidance. With the collaboration of semantics-agnostic panoptic matching and region-biased selection and extension, our CPRAL can strike a balance between labeling efforts and performance and compromise the semantics distribution. We perform extensive experiments on Cityscapes and BDD10K datasets and show that CPRAL outperforms the cutting-edge methods with impressive results and less labeling proportion.

Summary Fault localization is sensitive to program runs, and the pattern of fault propagation and manifestation in real software is extremely complex and uncertain. To accommodate the complexity and uncertainty, this paper presents a novel probabilistic graph model – the probabilistic cause–effect graph (PCEG) is built upon dynamic dependencies generated from running the faulty program against failed test cases and performs probabilistic inference with coverage information from the whole test suite. PCEG is an extension of the traditional probabilistic graph both in structural and inferential terms and is different from earlier probabilistic approaches to software diagnosis by introducing two forms of evidences (i.e. apparent faults and real faults). The proposed probabilistic reasoning algorithm works on the PCEG converted from a dynamic program dependency graph and diagnoses the causes with both top‐down and bottom‐up inference. The experimental results have shown the improvements on diagnostic effectiveness and accuracy in both single‐fault and multiple‐fault context, even when a program yields similar program runs through loop statements. Copyright © 2015 John Wiley & Sons, Ltd.

High-performance SOCs (system on a chip) are being widely used in multimedia signal processing. Timing behavior is a critical issue in high-performance design, and one of the significant and challenging problems in the SOC design is to ensure the safety of timing behavior. A model that combines Allen-Givone algebra in multi-valued logic and the waveform polynomial is presented in this paper. Functional and timing behavior can be precisely described simultaneously by one unified representation, allowing inputs and outputs of a multi-valued logic function to be described by multi-valued waveforms, which are consistent with commonly used intuitive waveforms. This method is applicable in precisely analyzing and checking the correctness and safety of SOC timing attributes.