arXiv subjects

Tianjian Chen

Publications and source records attributed to Tianjian Chen.

22 records · Page 2Linked to original sources

arXiv2019

Federated Transfer Reinforcement Learning for Autonomous Driving

Reinforcement learning (RL) is widely used in autonomous driving tasks and training RL models typically involves in a multi-step process: pre-training RL models on simulators, uploading the pre-trained model to real-life robots, and fine-tuning the weight parameters on robot vehicles. This sequential process is extremely time-consuming and more importantly, knowledge from the fine-tuned model stays local and can not be re-used or leveraged collaboratively. To tackle this problem, we present an online federated RL transfer process for real-time knowledge extraction where all the participant agents make corresponding actions with the knowledge learned by others, even when they are acting in very different environments. To validate the effectiveness of the proposed approach, we constructed a real-life collision avoidance system with Microsoft Airsim simulator and NVIDIA JetsonTX2 car agents, which cooperatively learn from scratch to avoid collisions in indoor environment with obstacle objects. We demonstrate that with the proposed framework, the simulator car agents can transfer knowledge to the RC cars in real-time, with 27% increase in the average distance with obstacles and 42% decrease in the collision counts.

cs.LG↗

arXiv2019

Federated Machine Learning: Concept and Applications

Today's AI still faces two major challenges. One is that in most industries, data exists in the form of isolated islands. The other is the strengthening of data privacy and security. We propose a possible solution to these challenges: secure federated learning. Beyond the federated learning framework first proposed by Google in 2016, we introduce a comprehensive secure federated learning framework, which includes horizontal federated learning, vertical federated learning and federated transfer learning. We provide definitions, architectures and applications for the federated learning framework, and provide a comprehensive survey of existing works on this subject. In addition, we propose building data networks among organizations based on federated mechanisms as an effective solution to allow knowledge to be shared without compromising user privacy.

cs.AI↗

arXiv2018

Underactuated Hand Design Using Mechanically Realizable Manifolds

Hand synergies, or joint coordination patterns, have become an effective tool for achieving versatile robotic grasping with simple hands or planning algorithms. Here we propose a method to determine the hand synergies such that they can be physically implemented in an underactuated fashion. Given a kinematic hand model and a set of desired grasps, our algorithm optimizes a Mechanically Realizable Manifold designed to be achievable by a physical underactuation mechanism, enabling the resulting hand to achieve the desired grasps with few actuators. Furthermore, in contrast to existing methods for determining synergies which are only concerned with hand posture, our method explicitly optimizes the stability of the target grasps. We implement this method in the design of a three-finger single-actuator hand as an example, and evaluate its effectiveness numerically and experimentally.

cs.RO↗

arXiv2018

Proprioception-Based Grasping for Unknown Objects Using a Series-Elastic-Actuated Gripper

Grasping unknown objects has been an active research topic for decades. Approaches range from using various sensors (e.g. vision, tactile) to gain information about the object, to building passively compliant hands that react appropriately to contacts. In this paper, we focus on grasping unknown objects using proprioception (the combination of joint position and torque sensing). Our hypothesis is that proprioception alone can be the basis for versatile performance, including multiple types of grasps for objects with multiple shapes and sizes, and transitions between grasps. Using a series-elastic-actuated gripper, we propose a method for performing stable fingertip grasps for unknown objects with unknown contacts, formulated as multi-input-multi-output (MIMO) control. We also show that the proprioceptive gripper can perform enveloping grasps, as well as the transition from fingertip grasps to enveloping grasps.

cs.RO↗