A Sub- 10 μs In-Memory-Search Collision Detection Accelerator Based on RRAM-TCAMs
- 12 May 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 44 (12) , 4510-4523
Abstract
Collision detection can account for over 90% of the total calculation time in motion planning, which makes it critical to optimize the collision detection process. Previous hardware accelerators can hardly maintain fast computation speed for real-time collision detection while supporting a large roadmap. In this work, we present a novel in-memory-search accelerator. The accelerator achieves an impressive sub- $10~\mu $ s time for collision detection in 800-MB scale roadmaps with 100000 edges. Such accelerator leverages an in-memory-search architecture that can perform search operation in massively parallel manner with low energy consumption. It is powered by ternary content-addressable memories (TCAMs) based on large-scale nonvolatile resistive random-access memory (RRAM) arrays, which not only store a large roadmap on chip, perform look-up table function in a single clock cycle to accelerate collision detection by combining with the unique logic encoding scheme. Importantly, the RRAM-based TCAM eliminates the need for extensive data transfer between memory and computing units during collision detection, leading to significant energy and delay saving. As the pioneering TCAM-based in-memory-search accelerator for collision detection, the structure achieves 210– $2225\times $ times speedup compared with previous application-specific integrated circuit-based accelerators, and more than 33 $879\times $ times speedup compared with CPU-based collision detection. Such accelerator well exceeds the speed requirement for collision detection (<1 ms), and is suitable for a number of applications, such as robot motion planning in dynamic environment, animation, machining and manufacturing.
Keywords
Funding Information
- National Natural Science Foundation of China (92364107, 62104140, 62402311)
- National Key Research and Development Program of China (2024YFE0204300)
- Science and Technology Commission of Shanghai Municipality (23QA1405300, 24ZR1491500, 24ZR1433700)
- Natural Science Foundation of Chongqing Municipality (CSTB2022NSCQ-MSX1095)
- State Key Laboratory of Radio Frequency Heterogeneous Integration
This publication has 33 references indexed in Scilit:
- Accelerating Path Planning for Autonomous Driving with Hardware-Assisted Memorization2022
- Memory devices and applications for in-memory computingNature Nanotechnology, 2020
- Compute-in-Memory with Emerging Nonvolatile-Memories: Challenges and Prospects2020
- Resistive switching materials for information processingNature Reviews Materials, 2020
- Ferroelectric ternary content-addressable memory for one-shot learningNature Electronics, 2019
- A Programmable Architecture for Robot Motion Planning Acceleration2019
- Ternary content-addressable memory with MoS2 transistors for massively parallel data searchNature Electronics, 2019
- Dadu-P: A Scalable Accelerator for Robot Motion Planning in a Dynamic Environment2018
- OctoMap: an efficient probabilistic 3D mapping framework based on octreesAutonomous Robots, 2013
- Probabilistic roadmaps for path planning in high-dimensional configuration spacesIEEE Transactions on Robotics and Automation, 1996