Boxi: Design Decisions in the Context of Algorithmic Performance for Robotics

Abstract: Achieving robust autonomy in mobile robots operating in complex and unstructured environments requires a multimodal sensor suite capable of capturing diverse and complementary information. However, designing such a sensor suite involves multiple critical design decisions, such as sensor selection, component placement, thermal and power limitations, compute requirements, networking, synchronization, and calibration. While the importance of these key aspects is widely recognized, they are often overlooked in academia or retained as proprietary knowledge within large corporations. To improve this situation, we present Boxi, a tightly integrated sensor payload that enables robust autonomy of robots in the wild. This paper discusses the impact of payload design decisions made to optimize algorithmic performance for downstream tasks, specifically focusing on state estimation and mapping. Boxi is equipped with a variety of sensors: two LiDARs, 10 RGB cameras including high-dynamic range, global shutter, and rolling shutter models, an RGB-D camera, 7 inertial measurement units (IMUs) of varying precision, and a dual antenna RTK GNSS system. Our analysis shows that time synchronization, calibration, and sensor modality have a crucial impact on the state estimation performance. We frame this analysis in the context of cost considerations and environment-specific challenges. We also present a mobile sensor suite cookbook to serve as a comprehensive guideline, highlighting generalizable key design considerations and lessons learned during the development of Boxi. Finally, we demonstrate the versatility of Boxi being used in a variety of applications in real-world scenarios, contributing to robust autonomy. More details and code: https://github.com/leggedrobotics/grand_tour_box

Algorithmic Performance and Design Decisions in Robotics

This paper presents Boxi, a multi-modal sensor payload designed to enhance the autonomy of mobile robots operating in complex environments. The integration of various sensors and careful design decisions is explored in the context of algorithmic performance, particularly focusing on state estimation and mapping. The paper emphasizes the criticality of sensor modality, time synchronization, and extrinsic calibration on performance outcomes, presenting rigorous experimental evaluations in diverse real-world scenarios.

Key Findings and Observations

The payload is equipped with diverse sensor modalities, which include LiDARs, RGB cameras, IMUs, and a GNSS system, allowing for comprehensive perception. The paper provides several insights into the functioning and performance of these sensors:

• Sensor Modalities: LiDAR-based state estimation algorithms (DLIO) consistently outperform visual-inertial methods (OKVIS2) in terms of Absolute Trajectory Error (ATE), particularly in environments with ample geometric features. GNSS provides highly precise global positioning under conditions of good satellite visibility.
• Camera Performance: The effects of camera type on performance are notable, with global shutter cameras (CoreResearch) outperforming rolling shutter cameras (HDR, ZED2i) in visual-inertial odometry tasks. This highlights the importance of accounting for rolling shutter effects to optimize algorithmic outputs.
• IMU Analysis: The precision of IMU sensors is shown to impact state estimation performance marginally when tightly coupled with LiDAR due to robust point cloud registration processes. However, higher-grade IMUs demonstrate significant advantages in dead-reckoning scenarios during sensor measurement dropout periods.
• Synchronization and Calibration: Accurate time synchronization (preferably < 5 milliseconds) and extrinsic calibration are highlighted as pivotal for optimizing sensor fusion. Perturbations in these parameters lead to performance degradation, underscoring the need for precision in hardware setup and system integration.

Practical and Theoretical Implications

The research presented in this paper has significant implications for the design and deployment of sensor suites in robotic systems:

1. Algorithmic Performance Optimization: The identification of optimal sensor configurations and calibration procedures can lead to substantial improvements in the state estimation benchmarks critical to robotic autonomy.
2. Cost and Environment Considerations: Understanding sensor performance relative to environmental challenges can guide cost-effective solutions tailored for specific deployment conditions, particularly in extreme or dynamic environments.
3. Future Research Directions: The exploration of additional modalities, such as UWB or event-based sensors, could further enhance perception capabilities. The development of real-time processing algorithms to account for sensor offsets and rolling shutter effects is paramount.

Conclusion

This research elucidates the relationship between sensor payload design decisions and the algorithmic performance of state estimation and mapping in robotics. By providing detailed analysis and real-world testing, the paper offers valuable guidelines for developing robust autonomous systems. It presents a strong case for integrated design and development, calling for deeper investigation into sensor and algorithm interactions to realize the full potential of robotic autonomy.