Advanced Virtual Exercise Environment Device for Promoting Physical Activity

D3: Advanced Virtual Exercise Environment Device (AVEED) for Promoting Socially Engaging Physical Activity in People with Disabilities


RecTech has been developing a proof-of-concept advanced virtual exercise environment device (AVEED) and a utility patent has been filed.  AVEED is a universal, modular exercise device that can be taken into mass production at an affordable cost, which will allow people with disabilities to participate in more enjoyable and engaging exercise experiences.  The AVEED unit is unlike any other commercial product available on the market.  First, the machine features a universally adaptable positioning system capable of being used from a recumbent position to a forward lean position, which allows the device to be moved into the optimal spatial orientation to meet an individual user’s needs, either sitting or standing, and also meets a range of user function, height and limb length. Secondly, AVEED uses an advanced software control system that has enabled several unique capabilities to meet the varying demands of people with physical disabilities.  The latest advancements have centered around our ability to input unilimb loading resistance to both the left and right arms/legs independent of each.  Since the arm/leg cranks are independent, the device is uniquely suited for allowing a training regimen to be arranged where the weakened limb(s) and the stronger limb(s) are being loaded relatively equally so that a training effect can occur for each limb.  To promote socially engaging exercise, AVEED also features preliminary capabilities to interface with any virtual environment training (VET) system such as Bkool or Zwift.  Although these commercial VET systems have seen promising advancements in the past few years, these systems still pose an access problem for people with physical disabilities (i.e., non-ambulatory or limited ambulation).


In order for people with physical disabilities to be able to fully participate in publicly available VET, we are proposing to add the following features to AVEED: a) Power Augmentation Capabilities: to offer smart power augmentation1-3 by adding four different motors in place of the current flywheel-based resistance system.  This will allow users with uneven power on different limbs (hemiparesis)4-6 to derive maximum benefits of the cycling motion; b) Precise Measurement of Input Power: to gather real-time power output in watts from each input (left arm, right arm, left leg, right leg).  This will allow accurate measurement for more realistically interfacing with virtual exercise environments and to support clinical rehabilitation applications; c) Simplified Adjustment of Upper Axes Position: to advance the portions of the frame carrying the arm axes so they will rotate electronically about their respective pivot axes similar to the way you can easily adjust a powered seat back angle in a car; and d) Refine VET Connectivity: AVEED is currently able to connect to online VET systems such as Zwift, but the connection needs to be optimized for accurate input power calibration, adjusting between upper and lower extremity input, in addition to a simplified connection methodology.

We will also conduct an exploratory study to test the enhanced AVEED device with post-stroke participants. The aims of the study are described below.

  1. To use AVEED to discover the optimal resistance/power augmentation settings that will enable people post-stroke to generate the greatest arm and leg pedaling work values during exercise with the device.
  2. To use these optimal parameters obtained from Aim 1to inform a 12-week pilot and feasibility study.


Twenty-four individuals with chronic hemiparetic stroke will be recruited.  Inclusion criteria are: (1) age 18 years or older; (2) >5 months post-stroke; (3) able to ambulate at least 14 m with an assistive device or assistance of one person; (4) approval of primary care physician.  Exclusion criteria are: (1) Mini-Mental State Exam score <24; (2) currently receiving lower extremity strengthening exercises or gait training.

For Aim 1, individuals will participate in a 60-minute session where we will ask them to pedal with their arms, legs, and then both arms/legs using AVEED against different levels of resistance/power augmentation and with different left/right leg speeds.  We will collect data on heart rate, perceived exertion, and any challenges associated with the single session training protocol.  This optimization protocol will allow us to calculate the relative differences in pedaling work generated by each limb.  After safety and optimal parameters are ascertained, VET capabilities will be initiated.  We will collect feedback about their experience using the AVEED device while generating the greatest possible symmetry of work done by the paretic versus non-paretic limbs.  These generated optimal settings will be used in Aim 2.

For Aim 2, we will conduct a randomized crossover trial (12 participants per arm) with a 4-week washout period. Participants will be randomized to start with 30 minutes of AVEED multi-limb exercise training, three days per week, over a 12-week period, or be instructed on a daily walking regimen for the same time frame.  With respect to AVEED training, the participants will pedal for 30 minutes at a pace and against resistance levels that elicit the greatest symmetry of pedal work with each limb and a heart rate response between 50% and 80% of their heart rate reserve.  The walking group will be asked to walk around Lakeshore’s indoor track for 30 minutes continuously, three days a week, at their own comfortable pace. Process feasibility outcomes will include recruitment, refusal, retention and adherence. Clinical outcomes of interest will be assessed pre and post intervention.  Measures will include:

  1. 10-meter timed walk: Gait speed (m/sec) = distance covered (10 m)/time (sec)
  2. 6-minute walk test: Distance covered to the nearest meter in 6 min
  3. Berg Balance Scale: 14 items (1 sitting, 13 standing) related to balance function

We will also collect biomechanical muscle performance data to explore possible mechanistic underpinnings of the changes that are observed pre- versus post-training.  Changes to peak ankle plantarflexor and peak hip flexor torque will be assessed using a multi-joint dynamometer (Biodex System 3, Shirley, NY).

Final Outcomes

The enhanced AVEED will allow users with hemiparesis to generate the greatest arm and leg pedaling work during exercise with the device, which will be supported by the 12-week pilot and feasibility study. The device will be prepared for commercialization.


  1. Kazerooni H. Exoskeletons for human power augmentation. Paper presented at: Intelligent Robots and Systems, 2005.(IROS 2005). 2005 IEEE/RSJ International Conference on2005.
  2. Marcheschi S, Salsedo F, Fontana M, Bergamasco M. Body extender: whole body exoskeleton for human power augmentation. Paper presented at: Robotics and Automation (ICRA), 2011 IEEE International Conference on2011
  3. Yang C-J, Niu B, Chen Y. Adaptive neuro-fuzzy control based development of a wearable exoskeleton leg for human walking power augmentation. Paper presented at: Advanced Intelligent Mechatronics. Proceedings, 2005 IEEE/ASME International Conference on2005.
  4. Chen G, Patten C, Kothari DH, Zajac FE. Gait differences between individuals with post-stroke hemiparesis and non-disabled controls at matched speeds. Gait & posture. 2005;22(1):51-56
  5. Levin MF, Cirstea MC, Michaelsen SM, Roby-Brami A. Use of trunk for reaching targets placed within and beyond the reach in adult hemiparesis. Exp Brain Res. 2002;143.
  6. Stevens JA, Stoykov MEP. Using motor imagery in the rehabilitation of hemiparesis. Archives of physical medicine and rehabilitation. 2003;84(7):1090-1092