Biped humanoid robots hold the potential to perform human-like locomotion and manipulation tasks in a variety of situations where either skilled humans are unavailable or where environmental conditions prevent human intervention. However, while bipedal motion maximizes maneuverability in tight workspaces, it simultaneously reduces locomotion stability. To initiate a new direction of ongoing research in legged robot motion control, this work will enable humanoid robots to transform into tripeds or quadrupeds or, more generally, "SupraPeds". To control the potentially numerous contact points on SupraPeds, we will also develop a software system that implements generic multi-contact control for arbitrary humanoids, which will enable autonomous balancing while satisfying contact force constraints.
This concept is intuitive to humans as everyone has experienced the need for an extra hand or two to maintain balance or navigate difficult terrain. The PIs will capitalize on this aspect to communicate to high school students, undergraduates, and graduate students, as well as educators nationwide, through live and online courses and seminars. To broaden the participation of under-represented groups, we will participate in a Research Experience for Undergraduates (REU) program especially for community colleges and minority-serving institutions. The results of the research will be published in open-access conferences and journals whenever possible, and the sensing and control software will be documented and published on public websites as open-source code. The applications of the work will impact fields such as search and rescue and scientific exploration.
The challenges in deploying humanoid robots in unstructured environments are numerous, and range from being able to traverse rough terrain, integrate information from tactile, kinematic, and vision sensors, execute complex manipulation, and gracefully compensate for sensor occlusion and actuator limits. Efforts to make robots emulate humans, however, have ignored an important fact: whenever humans approach their limits, they augment their capabilities with a diverse variety of tools. Perhaps the most useful and easily adopted tool is the walking staff, which improves support, enables load redistribution to the upper body, and can also be used as a sensor to probe the stability of planned footsteps. In this project, we proposed the concept of SupraPed platform as shown in Fig. 1, a set of smart poles, a suite of multi-contact control model, and a haptic tele-operation system, that transforms biped humanoid robots into tripeds or quadrupeds or more generally, SupraPeds. Quadrupeds surpass bipeds in distributing their weight over more actuators and thus increasing their payload. In addition, their extra limbs expand their support region and improve stability while traversing uneven terrain. Bipeds, on the other hand, excel at navigating through narrow passages and at maximizing their available workspace while manipulating objects. Most unstructured environments, however, require a combination of both, which makes the SupraPed platform ideal for augmenting humanoid robots. Fig 2 illustrates the entire framework of SupraPeds. To synthesize complex behaviors of humanoid robots, a friction-consistent whole-body control framework, is developed to integrate manipulation, locomotion, and diverse dynamic constraints such as multi-contact interaction, friction property, and joint limits. This framework integrates task-oriented dynamic control and control prioritization, allowing robots to control multiple task primitives while complying with physical and movement-related constraints. This framework implements generic multi-contact control for arbitrary humanoids and enables autonomous balancing without violating friction constraints. The simulation results shown in the published paper demonstrated that the integration of the SupraPed platform and the proposed control framework signi?cantly enhance the locomotion performance of humanoid robots in unstructured rough-terrain environments (Fig. 3). Moreover, in this project, a prototype of the smart pole was designed and fabricated in the lab. Fig. 4 displays the design details and actual implementation. We developed a novel pole extension design with mechanical multiplexing mechanism, shown in Fig. 5, capable of resisting 500N axial force, 100N lateral force and 20 N-m bending moment applied on the pole’s tip. To satisfy the sensing force range and limited space requirement, a novel 5Dof F/T sensor is invented for the smart pole (Fig. 6). Unlike the conventional approach, our sensor design decouples the axial force and bending moments, and is able to achieve the same performance with the smaller structure. In brief, this research work proposed to enable humanoid robots to operate in complex unstructured environments by developing a set of actuated smart staffs with detailed local perception, which will transform biped humanoids into tripeds or quadrupeds. To control numerous contact points on SupraPeds, we developed a software system that implements generic multi-contact control for arbitrary humanoids, which will enable autonomous balancing while satisfying contact force constraints. The first prototype of smart staff has been designed, fabricated and tested in the lab as well. All mechanical designs and source code resulting from this project will be published and made freely available online on the project’s website helping other researchers to get detailed insights or even replicate the SupraPeds platform. Although research on robot mechanisms and algorithms for rescue robots working in hazardous environments has made tremendous progress over the past decade, we still believe that SupraPeds will bring a whole new level of mobility and manipulation for humanoid robots.
