Revolutionizing Prosthetics is an ambitious multiyear program—funded by the Defense Advanced Research Projects Agency (DARPA)—to create a neurally controlled artificial limb that will restore near-natural motor and sensory capability to upper-extremity amputee patients. APL is leading an interdisciplinary team consisting of other Johns Hopkins institutions, government agencies, universities, and private firms to implement DARPA’s vision of providing the most advanced upper-extremity prosthesis.
APL has developed novel intelligent systems technologies that enable a team of robots to autonomously observe, orient, decide, and act while interfacing with human “teammates” on critical applications. Using our marsupial team-planning algorithms, a lead robot can quarterback ground and air robots through a cluttered indoor space while relaying information about the surroundings—and even opening doors. Using APL-developed algorithms, these robots can offer descriptions of novel objects in their camera feed, such as “dangerous” or “electronic device.” Integrating these technologies has allowed us to demonstrate high-level command of the robot team by a remote operator through goal indication in a previously unexplored area on a map and interfacing through voice queries for situational reports.
Solid-State Thermoelectric Technologies
Under the Defense Advanced Research Projects Agency’s Materials for Transduction effort, we lead an industry–academia–Department of Defense team working on novel coolers integrated with nanoengineered materials. Building on state-of-the-art capabilities with thin-film thermoelectrics for near-room-temperature cooling, sensing, and harvesting energy, we also are leveraging capabilities for other advanced thermoelectric device applications—including the Operational Lightweight Intelligent Thermo Electric power source for mobile platforms as well as a project to reduce auxiliary power units by using natural heat sources in hypersonic platforms.
Autonomy Testing and Evaluation
We are developing methods and procedures to ensure the safety and performance of autonomous systems. The Safe Testing of Autonomy in Complex, Interactive Environments (TACE) program provides two core testing and evaluation capabilities: first is a watchdog function consisting of onboard hardware and software that will take control of the test system should an action (like a change in climb rate or speed) violate the test parameters; second is a live-virtual-constructive environment in which the system’s autonomy can be replicated in combinations of live and virtual agents interacting in complete, realistic consideration of each other. We transitioned TACE technologies to the Department of Defense’s Atlantic Test Ranges and integrated with its range systems to enable safe transit of unmanned aerial systems to the ranges’ offshore test airspace without need for chase aircraft. The full TACE complement of capabilities is currently being transitioned to the Air Force Test Center at Edwards Air Force Base to enable a small unmanned aerial system infrastructure to test autonomy technologies.
Under the Intelligence Advanced Research Projects Activity’s HFGeo initiative, we are using science to accurately locate high-frequency targets refracted by the ionosphere. APL and project partner University of Bath developed techniques to analyze field-test data to estimate whether a “good” ionospheric model could meet program metrics—and based on analysis so far, it appears that it can. Furthermore, APL and Bath have developed research models that mitigate the effect of a dynamic ionosphere on angle-of-arrival observations exceeding program metrics.