Neuromorphic computing chip with spatiotemporal elasticity for multi-intelligent-tasking robots

Abstract

Recent advances in artificial intelligence have enhanced the abilities of mobile robots in dealing with complex and dynamic scenarios. However, to enable computationally intensive algorithms to be executed locally in multitask robots with low latency and high efficiency, innovations in computing hardware are required. Here, we report TianjicX, a neuromorphic computing hardware that can support true concurrent execution of multiple cross-computing-paradigm neural network (NN) models with various coordination manners for robotics. With spatiotemporal elasticity, TianjicX can support adaptive allocation of computing resources and scheduling of execution time for each task. Key to this approach is a high-level model, “Rivulet,” which bridges the gap between robotic-level requirements and hardware implementations. It abstracts the execution of NN tasks through distribution of static data and streaming of dynamic data to form the basic activity context, adopts time and space slices to achieve elastic resource allocation for each activity, and performs configurable hybrid synchronous-asynchronous grouping. Thereby, Rivulet is capable of supporting independent and interactive execution. Building on Rivulet with hardware design for realizing spatiotemporal elasticity, a 28-nanometer TianjicX neuromorphic chip with event-driven, high parallelism, low latency, and low power was developed. Using a single TianjicX chip and a specially developed compiler stack, we built a multi-intelligent-tasking mobile robot, Tianjicat, to perform a cat-and-mouse game. Multiple tasks, including sound recognition and tracking, object recognition, obstacle avoidance, and decision-making, can be concurrently executed. Compared with NVIDIA Jetson TX2, latency is substantially reduced by 79.09 times, and dynamic power is reduced by 50.66%.

Brain Inspired Computing: A Systematic Survey and Future Trends

“…For example, an HNN is able to achieve high accuracy from ANNs and rich dynamics, high efficiency, high robustness from SNNs [71], [238], [284]. The Tianjic team conducts many investigations on cross-paradigm comparison [71], [238], neural modeling [283], learning algorithms [263], hardware platforms [85], [86], and applications [285], [286] to build the ecology for such kind of hybrid BIC route. Recently, the hybridization idea has been broadly borrowed by the latest generation of BIC chips such as BrainScale 2 [82], SpiNNaker 2 [95], Loihi 2 [106], and the chips in [282], [287].…”

Brain-Inspired Remote Sensing Interpretation: A Comprehensive Survey

“…However, the computing power and extremely low power consumption of the neuromorphic computing chips are more worth looking forward. For example, TianjicX [333] has realized the experiment of a cat-and-mouse game under the condition of ultra-low power consumption and low delay. The total dynamic power consumption of the chip in the experiment is only 0.6w.…”

A review: Machine learning for strain sensor-integrated soft robots

“…That report well filled the gap of selfsupervised model training and updating during continuous usage, enabling optimal model performance under long-term operations. On the other hand, S. Ma et al developed a neuromorphic computing chip of TianjicX, which enabled multiple neural network models to be executed locally (Ma et al, 2022). As shown in Figure 1H, by equipping with TianjicX and multimodal sensors, without steaming sensor data out and external computing devices, a robotic car is able to simultaneously process multisensory information locally for diverse tasks, including object recognition, obstacle avoidance, and decision-making.…”

“…By equipping with TianjicX and multimodal sensors, a robotic car is able to simultaneously process multisensory information locally for various tasks including object detection, obstacle avoidance, and decisionmaking. Reproduced with permission fromMa et al (2022). Copyright: American Association for the Advancement of Science.…”