With the rapid expansion of autonomous robotic systems in recent years, humanoid robots are also gaining considerable attention. However, their motion control presents more complex challenges than wheeled mobile robots. For the first time, this work presents a complete footstep planning SoC chip for humanoid robots. It includes a novel time-domain graph search engine for 3D footstep planning, along with a mixed-signal zero-moment point (ZMP) gait scheduler enhanced by neural inverse kinematics for efficient motion control. This work is demonstrated in-situ on a fully assembled robot using the 65nm system-on-chip (SoC), achieving the best-in-class performance and power including an 18.4x enhancement on energy efficiency for motion control and a 2.7x energy saving in graph search compared to previous works. The demo video is provided in https://youtu.be/kBe-aRnzmG4.

This paper focuses on the precise identification of potential cloud computer users utilizing telecom big data. By integrating multi-source heterogeneous data and leveraging key technical capabilities, including raw bitstream network data parsing and the construction of an innovative O-domain hierarchical tagging architecture, we develop a multidimensional user behavior profiling model. This model enables accurate identification of cloud computer usage patterns for both enterprise and individual users. Key innovations include: (1) a novel O-domain tagging framework that substantially enhances scenario classification accuracy; (2) an attentivestacking fusion model that dynamically prioritizes telecomspecific behavioral features to optimize prediction performance; and (3) a validated analysis plan demonstrating superior conversion outcomes and reduced operational costs in real-world deployments. The proposed attentive-stacking fusion model, trained on behavioral characteristics unique to telecom scenarios, significantly enhances the prediction accuracy of potential cloud computer user groups. Comparative experiments confirm that the analysis plan efficiently identifies high-conversion-potential cloud computer user segments, effectively addressing the challenge of high customer acquisition costs inherent in traditional marketing methods. This research establishes a novel pathway for telecom operators to leverage big data for targeted user mining, marketing expenditure reduction, and conversion efficiency improvement.

Tropical algebra solves complex problems with only sum and min/max operations replacing expensive multiplication and addition in linear algebra. Due to the low computing cost, tropical algebra has recently gained significant attention in a broad range of areas such as combinatorial optimization, scheduling, machine learning, etc. In this paper, we propose a generic hardware accelerator architecture for tropical algebra supporting a wide range of applications. Novel time-domain (TD) computing accelerators with special mapping, precision expansion and, unrolling techniques are proposed to further improve hardware efficiency. Test results on various tropical calculations including linear regression, dynamic programming, and neural network are shown to demonstrate an energy saving from 1.5X to 2.1X, latency saving from 2.6X to 5.2X, or an overall energy-delay-product (EDP) improvement from 3.9X-10.5X compared with conventional digital implementation manifesting the promise of the algebraic solution on low power edge devices.