Search

It is challenging for observing superbunching effect with true chaotic light, here we propose and demonstrate a method to achieve superbunching effect of the degree of second-order coherence is 2.42 with broadband stationary chaotic light based on a cascaded Michelson interferometer (CMI), exceeding the theoretical upper limit of 2 for the two-photon bunching effect of chaotic light. The superbunching correlation peak is measured with an ultrafast two-photon absorption detector which the full width at half maximum reaches about 95 fs. Two-photon superbunching theory in a CMI is developed to interpret the effect and is in agreement with experimental results. The theory also predicts that the degree of second-order coherence can be much greater than $2$ if chaotic light propagates $N$ times in a CMI. Finally, a new type of weak signals detection setup which employs broadband chaotic light circulating in a CMI is proposed. Theoretically, it can increase the detection sensitivity of weak signals 79 times after the chaotic light circulating 100 times in the CMI., Comment: 8 pages, 5 figures

General contact-rich manipulation problems are long-standing challenges in robotics due to the difficulty of understanding complicated contact physics. Deep reinforcement learning (RL) has shown great potential in solving robot manipulation tasks. However, existing RL policies have limited adaptability to environments with diverse dynamics properties, which is pivotal in solving many contact-rich manipulation tasks. In this work, we propose Contact-aware Online COntext Inference (COCOI), a deep RL method that encodes a context embedding of dynamics properties online using contact-rich interactions. We study this method based on a novel and challenging non-planar pushing task, where the robot uses a monocular camera image and wrist force torque sensor reading to push an object to a goal location while keeping it upright. We run extensive experiments to demonstrate the capability of COCOI in a wide range of settings and dynamics properties in simulation, and also in a sim-to-real transfer scenario on a real robot (Video: https://youtu.be/nrmJYksh1Kc)

The success of deep reinforcement learning (RL) and imitation learning (IL) in vision-based robotic manipulation typically hinges on the expense of large scale data collection. With simulation, data to train a policy can be collected efficiently at scale, but the visual gap between sim and real makes deployment in the real world difficult. We introduce RetinaGAN, a generative adversarial network (GAN) approach to adapt simulated images to realistic ones with object-detection consistency. RetinaGAN is trained in an unsupervised manner without task loss dependencies, and preserves general object structure and texture in adapted images. We evaluate our method on three real world tasks: grasping, pushing, and door opening. RetinaGAN improves upon the performance of prior sim-to-real methods for RL-based object instance grasping and continues to be effective even in the limited data regime. When applied to a pushing task in a similar visual domain, RetinaGAN demonstrates transfer with no additional real data requirements. We also show our method bridges the visual gap for a novel door opening task using imitation learning in a new visual domain. Visit the project website at https://retinagan.github.io/, Comment: International Conference on Robotics and Automation (ICRA) 2021

Joint-event-extraction, which extracts structural information (i.e., entities or triggers of events) from unstructured real-world corpora, has attracted more and more research attention in natural language processing. Most existing works do not fully address the sparse co-occurrence relationships between entities and triggers, which loses this important information and thus deteriorates the extraction performance. To mitigate this issue, we first define the joint-event-extraction as a sequence-to-sequence labeling task with a tag set composed of tags of triggers and entities. Then, to incorporate the missing information in the aforementioned co-occurrence relationships, we propose a Cross-Supervised Mechanism (CSM) to alternately supervise the extraction of either triggers or entities based on the type distribution of each other. Moreover, since the connected entities and triggers naturally form a heterogeneous information network (HIN), we leverage the latent pattern along meta-paths for a given corpus to further improve the performance of our proposed method. To verify the effectiveness of our proposed method, we conduct extensive experiments on four real-world datasets as well as compare our method with state-of-the-art methods. Empirical results and analysis show that our approach outperforms the state-of-the-art methods in both entity and trigger extraction., Comment: Accepted by ICPR 2020

Reinforcement learning methods have been developed to achieve great success in training control policies in various automation tasks. However, a main challenge of the wider application of reinforcement learning in practical automation is that the training process is hard and the pretrained policy networks are hardly reusable in other similar cases. To address this problem, we propose the cascade attribute network (CAN), which utilizes its hierarchical structure to decompose a complicated control policy in terms of the requirement constraints, which we call attributes, encoded in the control tasks. We validated the effectiveness of our proposed method on two robot control scenarios with various add-on attributes. For some control tasks with more than one add-on attribute attribute, by directly assembling the attribute modules in cascade, the CAN can provide ideal control policies in a zero-shot manner., Comment: arXiv admin note: text overlap with arXiv:1711.09142

In this paper, we continue our prior work on using imitation learning (IL) and model free reinforcement learning (RL) to learn driving policies for autonomous driving in urban scenarios, by introducing a model based RL method to drive the autonomous vehicle in the Carla urban driving simulator. Although IL and model free RL methods have been proved to be capable of solving lots of challenging tasks, including playing video games, robots, and, in our prior work, urban driving, the low sample efficiency of such methods greatly limits their applications on actual autonomous driving. In this work, we developed a model based RL algorithm of guided policy search (GPS) for urban driving tasks. The algorithm iteratively learns a parameterized dynamic model to approximate the complex and interactive driving task, and optimizes the driving policy under the nonlinear approximate dynamic model. As a model based RL approach, when applied in urban autonomous driving, the GPS has the advantages of higher sample efficiency, better interpretability, and greater stability. We provide extensive experiments validating the effectiveness of the proposed method to learn robust driving policy for urban driving in Carla. We also compare the proposed method with other policy search and model free RL baselines, showing 100x better sample efficiency of the GPS based RL method, and also that the GPS based method can learn policies for harder tasks that the baseline methods can hardly learn.

Current autonomous driving systems are composed of a perception system and a decision system. Both of them are divided into multiple subsystems built up with lots of human heuristics. An end-to-end approach might clean up the system and avoid huge efforts of human engineering, as well as obtain better performance with increasing data and computation resources. Compared to the decision system, the perception system is more suitable to be designed in an end-to-end framework, since it does not require online driving exploration. In this paper, we propose a novel end-to-end approach for autonomous driving perception. A latent space is introduced to capture all relevant features useful for perception, which is learned through sequential latent representation learning. The learned end-to-end perception model is able to solve the detection, tracking, localization and mapping problems altogether with only minimum human engineering efforts and without storing any maps online. The proposed method is evaluated in a realistic urban driving simulator, with both camera image and lidar point cloud as sensor inputs. The codes and videos of this work are available at our github repo and project website., Comment: 8 pages, 10 figures, 2 tables

Recently, ghost imaging has been attracting attentions because its mechanism would lead to many applications inaccessible to conventional imaging methods. However, it is challenging for high contrast and high resolution imaging, due to its low signal-to-noise ratio (SNR) and the demand of high sampling rate in detection. To circumvent these challenges, we here propose a ghost imaging scheme that exploits Haar wavelets as illuminating patterns with a bi-frequency light projecting system and frequency-selecting single-pixel detectors. This method provides a theoretically 100% image contrast and high detection SNR, which reduces the requirement of high dynamic range of detectors, enabling high resolution ghost imaging. Moreover, it can highly reduce the sampling rate (far below Nyquist limit) for a sparse object by adaptively abandoning unnecessary patterns during the measurement. These characteristics are experimentally verified with a resolution of 512 times 512 and a sampling rate lower than 5%. A high-resolution (1000 times 1000 times 1000) 3D reconstruction of an object is also achieved from multi-angle images., Comment: 8 pages and 6 figures

Single-pixel imaging is suitable for low light level scenarios because a bucket detector is employed to maximally collect the light from an object. However, one of the challenges is its slow imaging speed, mainly due to the slow light modulation technique. We here demonstrate 1.4MHz video imaging based on computational ghost imaging with a RGB LED array having a full-range frame rate up to 100MHz. With this method, the motion of a high speed propeller is observed. Moreover, by exploiting single-photon detectors to increase the detection efficiency, this method is developed for ultra-high-speed imaging under low light level.

Ghost imaging is usually based on optoelectronic process and eletronic computing. We here propose a new ghost imaging scheme, which avoids any optoelectronic or electronic process. Instead, the proposed scheme exploits all-optical correlation via the light-light interaction and the vision persistence effect to generate images observed by naked eyes. To realize high contrast naked-eye ghost imaging, a special pattern-scanning architecture on a low-speed light-modulation disk is designed, which also enables high-resolution imaging with lower-order Hadamard vectors and boosts the imaging speed. With this approach, we realize high-contrast real-time ghost imaging for moving colored objects.

Based on optical correlations, ghost imaging is usually reconstructed by computer algorithm from the acquired data. We here proposed an alternatively high contrast naked-eye ghost imaging scheme which avoids computer algorithm processing. Instead, the proposed scheme uses a photoelectric feedback loop to realize the multiplication process of traditional ghost imaging. Meanwhile, it exploits the vision persistence effect to implement integral process and to generate negative images observed by naked eyes. To realize high contrast naked-eye ghost imaging, a special pattern-scanning architecture on a low-speed light-modulation mask is designed, which enables high-resolution imaging with lower-order Hadamard vectors and boosts the imaging speed as well. Moreover, two kinds of feedback circuits, the digital circuit and the analog circuit, are presented respectively, which can achieve high-speed feedback operation on the light intensity. With this approach, we demonstrate high-contrast real-time imaging for moving objects.

Although deep reinforcement learning (deep RL) methods have lots of strengths that are favorable if applied to autonomous driving, real deep RL applications in autonomous driving have been slowed down by the modeling gap between the source (training) domain and the target (deployment) domain. Unlike current policy transfer approaches, which generally limit to the usage of uninterpretable neural network representations as the transferred features, we propose to transfer concrete kinematic quantities in autonomous driving. The proposed robust-control-based (RC) generic transfer architecture, which we call RL-RC, incorporates a transferable hierarchical RL trajectory planner and a robust tracking controller based on disturbance observer (DOB). The deep RL policies trained with known nominal dynamics model are transfered directly to the target domain, DOB-based robust tracking control is applied to tackle the modeling gap including the vehicle dynamics errors and the external disturbances such as side forces. We provide simulations validating the capability of the proposed method to achieve zero-shot transfer across multiple driving scenarios such as lane keeping, lane changing and obstacle avoidance., Comment: Published at IEEE ITSC 2018

Three-photon superbunching is experimentally observed with the recently proposed superbunching pseudothermal light. To the best of our knowledge, it is the first time that three-photon superbunching is observed in linear optical system. From the quantum optics point of view, three-photon superbunching is interpreted as the result of constructive-destructive three-photon interference. The key to observe three-photon superbunching in superbunching pseudothermal light is that all the different ways to trigger a three-photon coincidence count are in principle indistinguishable, which is experimentally guaranteed by putting a pinhole before each rotating groundglass to ensure that all the passed photons are within one coherence area. The observed three-photon superbunching is helpful to increase the visibility of ghost imaging and understand the physics of third-order interference of light., Comment: 4 pages, 3 figures. Submitted for publications