Behavioral cloning from human demonstrations has succeeded in programming a robot to generate fine-grained motion, but it is still challenging to learn multimodal trajectories such as with various speeds. This restricts the use of a robot dataset collected by multiusers because the different proficiency of robot operators makes the dataset have diverse distributions of speed. To tackle this issue, we develop Hierarchical Action Chunking Transformer with Vector-quantization (HACT-Vq) to efficiently learn temporal multimodality in addition to fine-grained motion. The proposed hierarchical model consists of a high-level policy to make planning for a latent subgoal and style, and a low-level policy to predict an action chunk conditioned with the latent subgoal and style. The latent subgoal and style are trained as discrete representations so that high-level policy can efficiently learn multimodal distributions of demonstrations and retrieve the mode of fast behavior. In experiments, we set up bimanual robots in both simulation and real-world environments, and collected demonstrations with various speeds. The proposed model with the quantized subgoal and style showed the highest success rates with fast imitation behavior. Our code is available at https://github.com/SamsungLabs/hierarchical-act.

Pose estimation is a key challenge in robot manipulation and grasping task. Current object pose estimation approaches based on 3D models and depth sensor information have difficulties to handle transparent objects because of the limitation to capture the accurate depth information. To address these issues, we present a 6DoF pose estimation approach, called GhostPose, which utilizes a novel 3D bounding box prediction network and multi-view geometry with cameras on manipulator robot. Our 3D bounding box prediction network is simple and light-weight by adding a small branch to a one-stage object detector. The network detects 2D projections of 3D bounding box vertices. Then, 3D points are reconstructed from the 2D results of the multiple viewpoints with camera motion information, i.e. extrinsic parameters, calculated from the robot joint angles. We also present generalized pose definition to address pose ambiguity of symmetric objects and keep consistency of geometric properties around feature points across both of the symmetric and asymmetric objects. Comparing with the previous pose estimation approaches, GhostPose is more generalized to environments and object types, because it does not require 3D models, object specific key points, predefined stereo settings and depth map. In experiments, it outperforms a state-of-the-art approach and shows generalized properties by applying to a real manipulator robot grasping system.

Integrated GPUs (iGPUs) are ubiquitous in today's client devices such as laptops and desktops. Examples include Intel's HD or Iris Graphics and AMD's APUs. An iGPU resides on the same chip as the CPU, which is in contrast to a conventional discrete GPU that would typically be connected over the PCI-E bus. Much like discrete GPUs, iGPUs are also capable of general purpose computation in addition to traditional graphics roles. Further, iGPUs have some interesting differences compared to traditional GPUs such as a cache-coherent memory hierarchy and a shared last level cache with the CPU. Despite their wide spread use, they are not studied very extensively. To the best of our knowledge, this paper introduces the first open source trace generation and microarchitectural simulation framework for Intel's integrated GPUs. We characterise the performance of Intel's Skylake and Kabylake GPUs through detailed microbenchmarks, and use the performance evaluations to guide our models and validate the simulator.

GPU architectures are increasingly important in the multi-core era due to their high number of parallel processors. Performance optimization for multi-core processors has been a challenge for programmers. Furthermore, optimizing for power consumption is even more difficult. Unfortunately, as a result of the high number of processors, the power consumption of many-core processors such as GPUs has increased significantly.

GPU architectures are increasingly important in the multi-core era due to their high number of parallel processors. Programming thousands of massively parallel threads is a big challenge for software engineers, but understanding the performance bottlenecks of those parallel programs on GPU architectures to improve application performance is even more difficult. Current approaches rely on programmers to tune their applications by exploiting the design space exhaustively without fully understanding the performance characteristics of their applications.

Heterogeneous multiprocessors are increasingly important in the multi-core era due to their potential for high performance and energy efficiency. In order for software to fully realize this potential, the step that maps computations to processing elements must be as automated as possible. However, the state-of-the-art approach is to rely on the programmer to specify this mapping manually and statically. This approach is not only labor intensive but also not adaptable to changes in runtime environments like problem sizes and hardware/software configurations. In this study, we propose adaptive mapping, a fully automatic technique to map computations to processing elements on a CPU+GPU machine. We have implemented it in our experimental heterogeneous programming system called Qilin. Our results show that, by judiciously distributing works over the CPU and GPU, automatic adaptive mapping achieves a 25% reduction in execution time and a 20% reduction in energy consumption than static mappings on average for a set of important computation benchmarks. We also demonstrate that our technique is able to adapt to changes in the input problem size and system configuration.