Navigating a nonholonomic robot in a cluttered, unknown environment requires accurate perception and precise motion control for real-time collision avoidance. This paper presents NeuPAN: a real-time, highly accurate, map-free, easy-to-deploy, and environment-invariant robot motion planner. Leveraging a tightly coupled perception-to-control framework, NeuPAN has two key innovations compared to existing approaches: 1) it directly maps raw point cloud data to a latent distance feature space for collision-free motion generation, avoiding error propagation from the perception to control pipeline; 2) it is interpretable from an end-to-end model-based learning perspective. The crux of NeuPAN is solving an end-to-end mathematical model with numerous point-level constraints using a plug-and-play (PnP) proximal alternating-minimization network (PAN), incorporating neurons in the loop. This allows NeuPAN to generate real-time, physically interpretable motions. It seamlessly integrates data and knowledge engines, and its network parameters can be fine-tuned via backpropagation. We evaluate NeuPAN on a ground mobile robot, a wheel-legged robot, and an autonomous vehicle, in extensive simulated and real-world environments. Results demonstrate that NeuPAN outperforms existing baselines in terms of accuracy, efficiency, robustness, and generalization capabilities across various environments, including the cluttered sandbox, office, corridor, and parking lot. We show that NeuPAN works well in unknown and unstructured environments with arbitrarily shaped objects, transforming impassable paths into passable ones.

Searching the targets of interest in large-scale remote sensing images is a fundamental problem, which becomes a very challenging issue when there is no relevant example at hand but a mental picture in mind. Hand-drawn sketch as a precise and convenient expression of the mental picture makes sketch-based remote sensing image retrieval (SBRSIR) an ideal choice to cope with this issue. However, the accuracy of SBRSIR algorithm, which is critical to effective retrieval, is still far behind classic query-by-example image retrieval. Two central limiting factors for this performance gap are: 1) the lack of effective cross-domain representations for bridging the domain gap between sketches and remote sensing images and 2) the absence of large sketch/remote sensing image data sets for developing, evaluating, and comparing the SBRSIR approaches. In this article, we first develop a novel SBRSIR model to learn a deep joint embedding space with discriminative losses, where adversarial training is used for the embedding space to learn domain-invariant representations. Then, we contribute a sketch/remote sensing image data set specifically for SBRSIR and provide a benchmark for subsequent researchers. Extensive experiments on the data set and large scene images demonstrate the effectiveness and superiority of the method for both the seen and unseen categories.

Event-based cameras are a new type of vision sensor that can encode spatial-temporal context in a pixel-level event stream. Its appealing properties offer great potential for applications requiring low processing latency and low power consumption. As an effective representation of events, the surface of active event (SAE) has become a favorable choice for corner detection and object classification, among others. These tasks apply normalizations as an essential preprocessing step to extract time-invariant features from SAEs. However, previous normalization methods have some drawbacks, including low efficiency, requiring parameter tuning, etc. These drawbacks largely limit their performances in practical tasks. In this work, we propose a highly efficient normalization method, i.e., chain normalization, to break the limits in the previous state-of-the-art. We leverage the inherent properties of SAE in designing. First, we propose a novel SAE implementation to utilize the characteristics of SAE. Compared with previous works, our method can efficiently capture the spatial and temporal relationships of events and enable robust normalization. Second, we further increase the efficiency by using a novel stacking strategy. We compare our method to the state-of-the-art with extensive experiments, showing high classification accuracy and a significant improvement in runtime performance. We also release the source codes for future distribution and improvement in the community.

Proper exposure settings are crucial for modern machine vision cameras to accurately convert light into clear images. However, traditional auto-exposure solutions are vulnerable to illumination changes, splitting the continuous acquisition of unsaturated images, which significantly degrades the overall performance of underlying intelligent systems. Here we present the neuromorphic exposure control (NEC) system. This system effectively alleviates the longstanding saturation problem at its core by exploiting bio-principles found in peripheral vision to compute a trilinear event double integral (TEDI). This approach enables accurate connections between events and frames in the physics space for swift irradiance prediction, ultimately facilitating rapid control parameter updates. Our experimental results demonstrate the remarkable efficiency, low latency, superior generalization capability, and bio-inspired nature of the NEC in delivering timely and robust neuromorphic synergy for lighting-robust machine vision across a wide range of real-world applications. These applications encompass autonomous driving, mixed-reality, and three-dimensional reconstruction.

Calibrating the extrinsic parameters of sensory devices is crucial for fusing multi-modal data. Recently, event cameras have emerged as a promising type of neuromorphic sensors, with many potential applications in fields such as mobile robotics and autonomous driving. When combined with LiDAR, they can provide more comprehensive information about the surrounding environment. Nonetheless, due to the distinctive representation of event cameras compared to traditional frame-based cameras, calibrating them with LiDAR presents a significant challenge. In this paper, we propose a novel method to calibrate the extrinsic parameters between a dyad of an event camera and a LiDAR without the need for a calibration board or other equipment. Our approach takes advantage of the fact that when an event camera is in motion, changes in reflectivity and geometric edges in the environment trigger numerous events, which can also be captured by LiDAR. Our proposed method leverages the edges extracted from events and point clouds and correlates them to estimate extrinsic parameters. Experimental results demonstrate that our proposed method is highly robust and effective in various scenes.

Bimodal objects, such as the checkerboard pattern used in camera calibration, markers for object tracking, and text on road signs, to name a few, are prevalent in our daily lives and serve as a visual form to embed information that can be easily recognized by vision systems. While binarization from intensity images is crucial for extracting the embedded information in the bimodal objects, few previous works consider the task of binarization of blurry images due to the relative motion between the vision sensor and the environment. The blurry images can result in a loss in the binarization quality and thus degrade the downstream applications where the vision system is in motion. Recently, neuromorphic cameras offer new capabilities for alleviating motion blur, but it is non-trivial to first deblur and then binarize the images in a real-time manner. In this work, we propose an event-based binary reconstruction method that leverages the prior knowledge of the bimodal target’s properties to perform inference independently in both event space and image space and merge the results from both domains to generate a sharp binary image. We also develop an efficient integration method to propagate this binary image to high frame rate binary video. Finally, we develop a novel method to naturally fuse events and images for unsupervised threshold identification. The proposed method is evaluated in publicly available and our collected data sequence, and shows the proposed method can outperform the SOTA methods to generate high frame rate binary video in real-time on CPU-only devices.