Embedded.fm · 2 Nov 2023 · 56m · Episode 463
Kevin Lannen, embedded systems engineer at LUCI Mobility, explains how their safety system integrates with Group 3 power wheelchairs to prevent accidents like curb drops and collisions. He details the sensor fusion strategy using Intel RealSense cameras, Texas Instruments radar, and ultrasonics to create a 360-degree point cloud, while emphasizing deterministic algorithms over machine learning due to power constraints. Lannen also discusses the regulatory path as an FDA Class I device, the challenge of reverse-engineering closed wheelchair control systems, and the importance of keeping the human user in control via override mechanisms.
early-stage companies
Other shows on the same ground, closest first.
Hosts Daniel and Forbode discuss two robotics projects: a pizza-making robot from MIT, Carnegie Mellon, and UCSD that uses a teacher-student AI framework to learn dough manipulation, and an autonomous wheelchair study by EPFL that highlights safety risks and the need for assisted navigation in crowds. They analyze the engineering challenges of soft robotics, sensor integration, and the legal implications of human-robot interaction.
Matt Watson of Opteran Technologies explains how reverse-engineering honeybee brains enables efficient, low-power collision avoidance and SLAM, contrasting this with the high compute demands of supervised deep learning. Dr. Perla Maiolino of the Oxford Robotics Institute discusses CySkin, a flexible, large-area artificial skin using capacitive triangular modules for tactile perception and manipulation. Dr. Dimitrios Kanoulas of UCL addresses the challenge of fusing multi-sensor data (LiDAR, cameras, IMUs) to handle uncertainty and real-time planning in dynamic, unstructured environments.
Keith Jackson, former CEO of On Semiconductor, discusses the company's strategy of acquiring over 20 companies, including AMI Semiconductor and Fairchild, to build a broad portfolio in analog, power, and image sensors for automotive and industrial markets. Jack Weast, Intel Mobileye’s VP for autonomous vehicle standards, explains the 'Responsibility Sensitive Safety' (RSS) mathematical model and its application in driver assistance systems, including the upcoming Geely deal. The episode also covers the upcoming IoT Security Virtual Conference, highlighting the need for 'Secure by Design' principles as over 70% of connected devices remain vulnerable.
Ryan McMichael, Director of Sensors and Systems at Zoox, discusses the company's purpose-built, driverless robotaxi designed for mobility-as-a-service. He details the sensor suite—including LIDAR, radar, cameras, thermal imaging, and microphones—and the strategy of using diversity and redundancy to ensure safety and uptime. McMichael also reflects on his transition from aerospace instrumentation at Johns Hopkins APL to automotive, highlighting the shift from low-volume, high-stakes systems to high-volume manufacturing with rigorous reliability standards.
EE Times editors report on CEA Letty's conference on edge AI, highlighting the European focus on local processing for data privacy and the debut of a Type 1 diabetes autonomous medical device. They also cover PCI-SIG's rapid development of Gen 6 PCI Express, noting the challenges of copper interconnects and board redesigns. Additionally, the episode discusses the hyper-fragmented sensor market at Sensors Expo and Colin Barnden's argument that Level 2+ driver assist systems are more viable for mass-market cars than fully autonomous vehicles.
The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Embedded.fm. The episode itself belongs to its publisher.