Energy-Efficient SLAM via Joint Design of Sensing, Communication, and Exploration Speed

TL;DR

The paper tackles energy efficiency in lifelong SLAM by jointly designing sensing, communication, and exploration speed for an edge-assisted SLAM pipeline using a robot with 2D LiDAR and odometry. It builds an end-to-end model of sensing, data fusion, wireless transmission, and DL-based map reconstruction, and shows that the optimal data-transfer ratio is $\rho^*=1$ with a sensing period $t_{\text{sens}}^* = \frac{4(L-2e)}{v N_D}$ and velocity $v^* = \frac{4(L-2e)}{T_{max}}$, reducing the problem to a two-parameter optimization over $t_{\text{sens}}$ and $v$. An upper bound $E_{total,up}(v)$ is derived to capture the trade-offs among mechanical, sensing, and communication energy, with derivative analysis indicating the minimum occurs at $v^*$. Experiments and simulations reveal distinct energy contributions: communication energy grows with area, motion energy increases roughly linearly, and LiDAR energy remains constant, illustrating the value of the proposed joint design for scalable, energy-aware lifelong SLAM in edge-enabled systems.

Abstract

To support future spatial machine intelligence applications, lifelong simultaneous localization and mapping (SLAM) has drawn significant attentions. SLAM is usually realized based on various types of mobile robots performing simultaneous and continuous sensing and communication. This paper focuses on analyzing the energy efficiency of robot operation for lifelong SLAM by jointly considering sensing, communication and mechanical factors. The system model is built based on a robot equipped with a 2D light detection and ranging (LiDAR) and an odometry. The cloud point raw data as well as the odometry data are wirelessly transmitted to data center where real-time map reconstruction is realized based on an unsupervised deep learning based method. The sensing duration, transmit power, transmit duration and exploration speed are jointly optimized to minimize the energy consumption. Simulations and experiments demonstrate the performance of our proposed method.

Figures (6)

• Figure 1: System schematics
• Figure 2: Timing diagram of robot operation
• Figure 3: DeepMapping structure
• Figure 4: Total energy versus sensing period length
• Figure 5: Energy consumption versus size of task area

Theorems & Definitions (3)

• Lemma 1
• proof
• proof