Autonomous Vehicle Compute Chassis Thermal and Vibration Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle hardware platforms lack the necessary compute power and thermal management to support advanced AI algorithms, and are not designed to fit compactly within smaller vehicle trunks while effectively cooling and synchronizing systems.
Innovation Solution
A compact hardware platform with integrated compartments for a control server, compute server, and IO subsystem, utilizing PCIe links for sensor data processing and Ethernet communication, along with a chassis design that includes air intake and exhaust for cooling and vibration reduction, to provide a high-performance computing environment within a vehicle trunk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If conventional small configuration models are used, then the device complexity is reduced, but the compute power is insufficient to support advanced AI algorithms
Solution Approach 1:
The system is divided into multiple independent compute modules (e.g., compute module 102, 104, 106) that can be individually configured and interconnected. Each module contains specialized processors (GPUs, FPGAs, ASICs) that can be selectively activated based on computational requirements, allowing the system to scale compute power without increasing overall system complexity.
Solution Approach 2:
The patent implements a hierarchical architecture where compute modules are nested within a modular chassis structure. The chassis contains standardized bays that can accommodate different types and numbers of compute modules, creating a nested configuration where smaller modular units are contained within a larger integrated system, enabling flexible scaling of compute power.
2Power
If multiple computers are used to host and interconnect over Ethernet, then the compute capacity is increased, but the volume of the hardware platform increases
Solution Approach 1:
Multiple compute modules that would traditionally require separate computer housings and Ethernet networking infrastructure are merged into a single integrated chassis. The modules share common power supply, cooling infrastructure, and physical housing, dramatically reducing the total volume required while maintaining high compute capacity through parallel processing across multiple modules.
Solution Approach 2:
The chassis is designed as a universal platform that can accommodate different types of compute modules (GPUs, FPGAs, ASICs) and support various computational workloads. This multi-functional design allows a single hardware platform to provide diverse compute capabilities without requiring multiple specialized systems, reducing overall volume.
3Temperature
If conventional cooling designs are used, then the device complexity is minimized, but the thermal management is insufficient for high-performance computing components
Solution Approach 1:
The cooling system is designed with localized cooling solutions for each compute module bay, allowing different thermal management strategies to be applied to different regions of the chassis based on heat generation patterns. High-performance compute modules with greater thermal output receive more aggressive cooling, while lower-power modules use passive or minimal cooling, optimizing thermal management without uniformly increasing system complexity.
4Volume of moving object
If the hardware platform is designed for compact form factor, then the volume is reduced to fit small car trunks, but the airflow for cooling is restricted
Solution Approach 1:
The cooling airflow is routed through vertical and three-dimensional pathways within the chassis structure, utilizing unused volumetric space. Air intake vents and exhaust channels are positioned to create efficient airflow patterns that move through the compute module bays without requiring increased external chassis dimensions, maintaining compact form factor while ensuring adequate cooling.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables efficient processing and control of autonomous vehicles by providing a compact, high-performance computing environment that fits within smaller vehicle trunks, ensuring adequate cooling and system synchronization, thereby enhancing the operational capabilities of autonomous driving systems.
Implementation Method 1
a chassis design that includes air intake and exhaust for cooling
Implementation Method 2
chassis design that includes air intake and exhaust for cooling and vibration reduction
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An apparatus includes a chassis housing a control server compartment, a compute server compartment, and an input and output (IO) subsystem compartment. The apparatus further includes an IO subsystem inserted into the IO subsystem compartment, a compute server inserted into the compute server compartment, and a control server inserted into the control server compartment coupled to the compute server via an Ethernet connection. The IO subsystem includes one or more IO modules, where at least some of the IO modules can be coupled to sensors. The compute server receives the sensor data from the IO subsystem via some PCIe links and generates planning and control data based on the sensor data for controlling the autonomous vehicle. The control server controls and operates the autonomous vehicle by sending control commands to hardware of the autonomous vehicle based on the planning and control data received from the compute server.